def main():
global preview
parse_args(sys.argv[1:])
cam = PiCamera()
cam.rotation = -90
cam.resolution = (width, height)
cam.framerate = 60
rawCapture = PiRGBArray(cam, size=(width, height))
startTime = time.time()
endTime = time.time() + capTime
frames = 0
for image in cam.capture_continuous(rawCapture, format="bgr", use_video_port=True):
if image is None:
print 'capture failed'
break
frames += 1
frame = image.array
if preview:
cv2.imshow("frame", frame)
key = cv2.waitKey(1) & 0xFF
if time.time() > endTime:
break
rawCapture.truncate(0)
print "Average Framerate for " + str(frames) + \
" frames was: " + str(float(frames) / capTime) + "fps"
cv2.destroyAllWindows()
def runCam():
# initialize the camera and grab a reference to the raw camera capture
camera = PiCamera()
camera.resolution = (640, 480)
camera.vflip = True
camera.framerate = 72
rawCapture = PiRGBArray(camera, size=(640, 480))
camera.start_preview()
time.sleep(0.1)
# capture frames from the camera
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
image = frame.array
# show the frame
cv2.imshow("Frame", image)
key = cv2.waitKey(1) & 0xFF
# clear the stream in preparation for the next frame
rawCapture.truncate(0)
# if the `q` key was pressed, break from the loop
if key == ord("q"):
break
class PiVideoStream: def __init__(self, resolution=(320, 240), framerate=32): # initialize the camera and stream self.camera = PiCamera() self.camera.resolution = resolution self.camera.framerate = framerate self.rawCapture = PiRGBArray(self.camera, size=resolution) self.stream = self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True) # initialize the frame and the variable used to indicate # if the thread should be stopped self.frame = None self.stopped = False def __init__(self, resolution=(320, 240), framerate=32, rotate=0): # initialize the camera and stream self.camera = PiCamera() self.camera.rotation = rotate self.camera.resolution = resolution self.camera.framerate = framerate self.rawCapture = PiRGBArray(self.camera, size=resolution) self.stream = self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True) # initialize the frame and the variable used to indicate # if the thread should be stopped self.frame = None self.stopped = False def start(self): # start the thread to read frames from the video stream t = Thread(target=self.update, args=()) t.daemon = True t.start() return self def update(self): # keep looping infinitely until the thread is stopped for f in self.stream: # grab the frame from the stream and clear the stream in # preparation for the next frame self.frame = f.array self.rawCapture.truncate(0) # if the thread indicator variable is set, stop the thread # and resource camera resources if self.stopped: self.stream.close() self.rawCapture.close() self.camera.close() return def read(self): # return the frame most recently read return self.frame def stop(self): # indicate that the thread should be stopped self.stopped = True
class CameraThread (threading.Thread):
def __init__(self):
threading.Thread.__init__(self)
# Camera setup
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
self.camera.vflip = True
self.camera.hflip = True
self.camera.video_stabilization = True
# Camera warmup
time.sleep(0.1)
# Camera configuration
self.camera.exposure_mode = 'off'
self.camera.awb_mode = 'off'
self.camera.awb_gains = 1.5
self.camera.iso = 100
self.camera.shutter_speed = 6660
def run(self):
global ALLSTOP
global target
global last_seen
global last_time
print "Camera thread start"
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# Grab frame
big = frame.array
image = cv2.resize(big, (res_var, int(big.shape[0]*res_var/big.shape[1])), interpolation = cv2.INTER_AREA)
image = image[30:127, :] #img[y:y+h, x:x+w]
# Cleanup
self.rawCapture.truncate(0)
# Get HSV image
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
# Color Filter
target.there, coords, target.where, image = ColorFilter2(image, hsv, [red_lower, blue_lower], [red_upper, blue_upper], (0,255,0), minArea)
if target.there:
if target.where < center_range and target.where > -center_range:
last_seen = "center"
elif target.where <= center_range:
last_seen = "right"
elif target.where >= -center_range:
last_seen = "left"
last_time = time.clock()
#print "Target", target.there, last_time, last_seen
if ALLSTOP:
break
def rpiCapture(opts):
global _frame
global _frameCounter
# initialize the camera and grab a reference to the raw camera capture
camera= PiCamera()
camera.resolution = CAPTURE_RES
camera.framerate = 4
rawCapture = PiRGBArray(camera, size=CAPTURE_RES)
# allow the camera to warmup
time.sleep(0.1)
# capture frames from the camera
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True) :
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
with _frameLock :
#copy frame handled by video capturing structure, so it won't be modified by next captures
_frame = numpy.copy(frame.array)
if(_frameCounter == 0) :
initStream(frame, opts)
_frameCounter += 1
# clear the stream in preparation for the next frame
rawCapture.truncate(0)
if _stopThreads :
return
class image_publisher:
def __init__(self):
self.camera = PiCamera()
self.camera.resolution = (1024, 768) #(1920, 1080)
self.camera.ISO = 100
self.camera.sa = 100
self.camera.awb = "flash"
self.camera.co = 100
self.rawCapture = PiRGBArray(self.camera)
time.sleep(0.1)
self.image_pub = rospy.Publisher("image_topic",Image)
self.small_image_pub = rospy.Publisher("small_image_topic",Image)
self.bridge = CvBridge()
def publishImage(self):
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
try:
cv_image = frame.array
except CvBridgeError as e:
print(e)
#cv_image = (cv_image * 0.5.astype(umpy.uint8)
self.rawCapture.truncate(0)
try:
self.image_pub.publish(self.bridge.cv2_to_imgmsg(cv_image, "bgr8"))
small = cv2.resize(cv_image, (0,0), fx=0.5, fy=0.5)
self.small_image_pub.publish(self.bridge.cv2_to_imgmsg(small, "bgr8"))
except CvBridgeError as e:
print("problem")
print(e)
class ThePiCamera(Camera):
"""
Class for Raspberry Pi camera frame retrieval.
"""
def __init__(self):
"""
Constructor for ThePiCamera. This initializes parameters for Raspberry
Pi camera capture.
"""
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
time.sleep(0.1) # allow the camera to warm up
def get_iterator(self):
"""
Returns an iterator for obtaining a continuous stream of camera frames.
"""
return self.camera.capture_continuous(self.rawCapture, format='bgr', use_video_port=True)
def get_frame(self, raw_frame):
"""
Retrieves the camera frame returned by the iterator, converted into a
2D array.
"""
array = raw_frame.array
self.rawCapture.truncate(0)
return array
def destroy(self):
"""
Cleans up memory used for Raspberry Pi camera capture.
"""
self.camera.close()
class VideoThread(Thread):
def __init__(self):
Thread.__init__(self)
self.camera = picamera.PiCamera()
self.camera.vflip = True
self.camera.framerate = 30
self.camera.resolution = (640, 480)
self.camera.use_video_port = True
self.camera.quality = 85
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
def run(self):
time.sleep(1) # warm up the camera
for frame in camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
image = frame.array
cv2.imshow("Frame", image)
key = cv2.waitKey(1) & 0xFF
self.rawCapture.truncate(0)
if key == ord("q"):
break
class FrameGrabber(threading.Thread):
def __init__(self, frameQueue, angleQueue):
threading.Thread.__init__(self)
self.frameQueue = frameQueue
self.angleQueue = angleQueue
self.camera = PiCamera()
self.camera.hflip = CAMERA_HFLIP
self.camera.vflip = CAMERA_VFLIP
self.camera.resolution = (CAMERA_WIDTH, CAMERA_HEIGHT)
self.camera.framerate = CAMERA_FRAMERATE
self.rawCapture = PiRGBArray(self.camera, size=(CAMERA_WIDTH, CAMERA_HEIGHT))
self.startCaptureTime = time.time()
self.sc = ServoController()
def run(self):
# allow the camera to warmup
runtime = time.time()
sleeptime = min(runtime - self.startCaptureTime,1)
time.sleep(sleeptime)
# capture frames from the camera
while not shutdownEvent.isSet():
angle = self.angleQueue.get(block=True)
self.sc.setCameraAngle(angle)
ts = time.time()
self.camera.capture(self.rawCapture, format='bgr', use_video_port=True)
self.frameQueue.put((ts,self.rawCapture.array,angle))
self.rawCapture.truncate(0)
class VideoCamera(object):
def __init__(self):
# Using OpenCV to capture from device 0. If you have trouble capturing
# from a webcam, comment the line below out and use a video file
# instead.
#self.video = cv2.VideoCapture(0)
self.camera = PiCamera()
#self.rawCapture = PiRGBArray(self.camera)
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
# allow the camera to warmup
time.sleep(0.1)
# If you decide to use video.mp4, you must have this file in the folder
# as the main.py.
# self.video = cv2.VideoCapture('video.mp4')
def __del__(self):
#self.video.release()
pass
def get_frame(self):
# grab an image from the camera
self.camera.capture(self.rawCapture, format="bgr")
image = self.rawCapture.array
self.rawCapture.truncate(0)
# We are using Motion JPEG, but OpenCV defaults to capture raw images,
# so we must encode it into JPEG in order to correctly display the
# video stream.
ret, jpeg = cv2.imencode('.jpg', image)
return jpeg.tostring()
def handle_get_bubblescope_properties(req):
with PiCamera() as camera:
# Camera settings
# I don't know what they mean
camera.resolution = (img_width,img_height)
camera.ISO = 100
camera.sa = 100
camera.awb = "flash"
camera.co = 100
raw_capture = PiRGBArray(camera)
time.sleep(0.1)
# Before doing anything, capture the first frame and find the inner circle
camera.capture(raw_capture, format="bgr")
image = raw_capture.array
image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
circles = cv2.HoughCircles(image_gray, cv2.cv.CV_HOUGH_GRADIENT, 1, 200, param1=50, param2=40, minRadius=10, maxRadius=180)
raw_capture.truncate(0)
if circles is not None:
x, y, r = circles[0][0]
res = GetBubblescopePropertiesResponse()
res.center = (x,y)
res.inner_radius = r*1.2
res.outer_radius = r*2.25
return res
else:
print "no circle found\n"
return GetBubblescopePropertiesResponse()
class VideoCamera(object):
def __init__(self):
# Using OpenCV to capture from device 0. If you have trouble capturing
# from a webcam, comment the line below out and use a video file
# instead.
# self.video = cv2.VideoCapture(0)
# If you decide to use video.mp4, you must have this file in the folder
# as the main.py.
# self.video = cv2.VideoCapture('video.mp4')
# --------------- ARW -------------
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
# Get a generator object that serves up the frames
self.frame_gen = self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True)
def __del__(self):
# self.video.release()
pass
def get_frame(self):
#success, image = self.video.read()
# We are using Motion JPEG, but OpenCV defaults to capture raw images,
# so we must encode it into JPEG in order to correctly display the
# video stream.
#ret, jpeg = cv2.imencode('.jpg', image)
#return jpeg.tobytes()
# --------------- ARW -------------
frame = self.frame_gen.next() # get next frame
image = frame.array
ret, jpeg = cv2.imencode('.jpeg', image ) # jpeg to buffer
self.rawCapture.truncate(0) # clear stream in prep for next frame
return jpeg.tobytes()
def video_thread():
# enable the thread to modify the global variable 'latest_video_frame': (this variable will be accessed by functions doing some sort of video analysis or video streaming)
global latest_video_frame
# create an instance of the RPI camera class:
camera = PiCamera()
# rotate the camera view 180 deg (I have the RPI camera mounted upside down):
camera.hflip = True
camera.vflip = True
# set resolution and frame rate:
camera.resolution = (640, 480)
camera.framerate = 30
# create a generator 'video_frame_generator' which will continuously capture video frames
# from the camera and save them one by one in the container 'generator_output': ('video_frame_generator' is an infinte iterator which on every iteration (every time 'next()' is called on it, like eg in a for loop) gets a video frame from the camera and saves it in 'generator_output'))
generator_output = PiRGBArray(camera, size=(640, 480))
video_frame_generator = camera.capture_continuous(generator_output, format="bgr", use_video_port=True)
# allow the camera to warm up:
time.sleep(0.1)
for item in video_frame_generator:
# get the numpy array representing the latest captured video frame from the camera
# and save it globally for everyone to access:
latest_video_frame = generator_output.array
# clear the output container so it can store the next frame in the next loop cycle:
# (please note that this is absolutely necessary!)
generator_output.truncate(0)
# delay for 0.033 sec (for ~ 30 Hz loop frequency):
time.sleep(0.033)
class CamThread(Thread): """ Thread to continuously read incoming frames from the raspberry pi camera and send those frames to all the clients connected to the server. """ def __init__(self): print 'init CamThread' self.cam = PiCamera() self.cam.resolution = (480,320) self.cam.framerate = 25 self.rawCap = PiRGBArray(self.cam) # Wait for the webcam to open sleep(0.2) self.sock = socket.socket(socket.AF_INET, socket.SOCK_DGRAM) Thread.__init__(self) self.daemon = True self.running = True def run(self): print 'CamThread running' for frame in self.cam.capture_continuous(self.rawCap, format='bgr', use_video_port=True): if not self.running: break img = self.rawCap.array rows, cols, channels = img.shape if rows > 0 and cols > 0: # 180 degrees rotation img = self.rotate(img, 180) # Serialize and send img_ser = self.encode(img, '.jpg') for c in connected_clients: self.send(img_ser, (c, CLIENT_PORT)) # Clear the camera buffer begore grabbing the next frame self.rawCap.truncate(0) self.sock.close() self.cam.close() def encode(self, img, ext='.jpg'): return cv2.imencode(ext, img)[1].tostring() def rotate(self, frame, angle): rows, cols, channels = frame.shape RotMat = cv2.getRotationMatrix2D((cols/2, rows/2), angle, 1) return cv2.warpAffine(frame, RotMat, (cols, rows)) def send(self, data, addr): # Send the data splitted in oackets of 1024 bytes while len(data) > 1024: self.sock.sendto(data[:1024], addr) data = data[1024:] if len(data) > 0: self.sock.sendto(data, addr) def stop(self): self.running = False
def make_photo():
timestamp = int(time())
camera.resolution = (1920,1080)
rawCapture = PiRGBArray(camera)
camera.capture('photo_'+str(timestamp)+'.png')
rawCapture.truncate(0)
camera.resolution = (640, 480)
rawCapture = PiRGBArray(camera)
class CameraThread (threading.Thread):
def __init__(self):
threading.Thread.__init__(self)
## Camera
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
# Camera warmup
time.sleep(0.1)
# Init red & blue threads
redThread = RedFilterThread(red_lower, red_upper, "red")
redThread.daemon = True
redThread.start()
blueThread = BlueFilterThread(blue_lower, blue_upper, "blue")
blueThread.daemon = True
blueThread.start()
def run(self):
global ALLSTOP
global targetPos
global hasTarget
global image
global hsv
print "Camera thread is running"
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# Grab frame
image = cv2.flip(frame.array,0)
image = cv2.resize(image, (100, int(image.shape[0]*100/image.shape[1])), interpolation = cv2.INTER_AREA)
# Get HSV image
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
# Cleanup
self.rawCapture.truncate(0)
# Compare
targetPosList = list(set(targetPos_red).intersection(targetPos_blue))
if len(targetPosList) > 0:
targetPos = targetPosList[0]
hasTarget = True
print "Match at", targetPos
else:
targetPos = 0
hasTarget = False
print "No match"
print "Red", targetPos_red
print "Blue", targetPos_blue
if ALLSTOP:
break
time.sleep(1)
class OpenCVTEST:
def __init__(self):
# initialize the camera and grab a reference to the raw camera capture
factor = 1.5
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
# allow the camera to warmup
time.sleep(0.1)
cv2.namedWindow("Video")
self.pictureBuffer = deque()
self.currentPicture = None
self.firstPicture = None
def showPicture(self, picture):
cv2.imshow("Video", picture)
def addPicture(self, picture):
self.pictureBuffer.append(picture)
def getCurrentPicture(self):
return self.currentPicture
def getPicture(self):
return self.pictureBuffer.popleft()
def forePlay(self):
number = 0
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
self.currentPicture = frame.array
self.addPicture(self.currentPicture)
self.rawCapture.truncate(0)
number += 1
if number >= 100:
break
self.play()
def play(self):
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
self.currentPicture = frame.array
self.addPicture(self.currentPicture)
self.showPicture(self.getPicture())
self.rawCapture.truncate(0)
key = cv2.waitKey(1) & 0xFF
if key == ord("q"):
break
print("Bye bye!")
def make_video():
timestamp = int(time())
camera.resolution = (1920,1080)
rawCapture = PiRGBArray(camera)
camera.start_recording('video_'+str(timestamp)+'.h264')
sleep(10)
camera.stop_recording()
rawCapture.truncate(0)
camera.resolution = (640, 480)
rawCapture = PiRGBArray(camera)
class PiVideoStream:
def __init__(self, resolution=(640, 480), framerate=48):
# initialize the camera and stream
self.camera = PiCamera()
#self.camera.video_stabilization = True
self.camera.resolution = resolution
self.camera.framerate = framerate
self.rawCapture = PiRGBArray(self.camera, size=resolution)
self.stream = self.camera.capture_continuous(self.rawCapture,
format="bgr", use_video_port=True)
# initialize the frame and the variable used to indicate
# if the thread should be stopped
self.frame = None
self.stopped = False
def start(self):
# start the thread to read frames from the video stream
t = Thread(target=self.update, args=())
t.daemon = True
t.start()
return self
def update(self):
# keep looping infinitely until the thread is stopped
for f in self.stream:
# grab the frame from the stream and clear the stream in
# preparation for the next frame
self.frame = f.array
self.rawCapture.truncate(0)
# if the thread indicator variable is set, stop the thread
# and resource camera resources
if self.stopped:
self.stream.close()
self.rawCapture.close()
self.camera.close()
return
def read(self):
# return the frame most recently read
return self.frame
def stop(self):
# indicate that the thread should be stopped
self.stopped = True
# Settings for consistent images capture
# Must be called some times after the camera is started
def consistent(self):
self.camera.shutter_speed = self.camera.exposure_speed
self.camera.exposure_mode = 'off'
g = self.camera.awb_gains
self.camera.awb_mode = 'off'
self.camera.awb_gains = g
def connect(self):
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
server_address = (conf.ip, conf.videoport)
sock.connect(server_address)
# initialize the camera and grab a reference to the raw camera capture
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 32
rawCapture = PiRGBArray(camera, size=(640, 480))
# allow the camera to warmup
time.sleep(0.1)
frm = 0
# capture frames from the camera
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
image = frame.array
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
gray = cv2.flip(gray,0)
gray = cv2.flip(gray,1)
frm = frm + 1
if (frm >= 256):
frm = 0
data = np.zeros((640), dtype=np.uint8)
data[0] = data[1] = data[2] = data[3] = data[5] = 32
sent = sock.sendto(data, server_address)
# for i in range(1,480):
# data = gray[i,:]
# sent = sock.sendto(data, server_address)
data = gray.reshape(640*480,1)
sent = sock.sendto(data, server_address)
#cv2.imshow("My Image", gray)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# clear the stream in preparation for the next frame
rawCapture.truncate(0)
if (self.keeprunning == False):
break
class PiVideoStream:
def __init__(self, resolution=(320, 240), format = "bgr", framerate=30):
# initialize the camera and stream
self.camera = PiCamera()
time.sleep(.2) #let camera init
self.camera.resolution = resolution
self.camera.framerate = framerate
self.camera.framerate = 30
self.camera.sharpness = -100
self.camera.contrast = 50
self.camera.saturation = 100
self.camera.brightness = 40
self.camera.awb_mode = 'off'
self.camera.awb_gains = (1.29, 1.44)
self.camera.shutter_speed = 800
self.camera.exposure_mode = 'night'
self.rawCapture = PiRGBArray(self.camera, size=resolution)
self.stream = self.camera.capture_continuous(self.rawCapture,
format=format, use_video_port=True)
time.sleep(.2) #let camera init
# initialize the frame and the variable used to indicate
# if the thread should be stopped
self.frame = None
self.stopped = False
def start(self):
# start the thread to read frames from the video stream
Thread(target=self.update, args=()).start()
return self
def update(self):
# keep looping infinitely until the thread is stopped
for f in self.stream:
# grab the frame from the stream and clear the stream in
# preparation for the next frame
self.frame = f.array
self.rawCapture.truncate(0)
# if the thread indicator variable is set, stop the thread
# and resource camera resources
if self.stopped:
self.stream.close()
self.rawCapture.close()
self.camera.close()
return
def read(self):
# return the frame most recently read
return self.frame
def stop(self):
# flag thread to stop
self.stopped = True
def run(self):
try:
with picamera.PiCamera() as camera:
camera.resolution = (self.CAMERA_WIDTH, self.CAMERA_HEIGHT)
camera.framerate = 32
rawCapture = PiRGBArray(camera, size=(self.CAMERA_WIDTH, self.CAMERA_HEIGHT))
time.sleep(0.1)
# Now fix the values
camera.shutter_speed = camera.exposure_speed
camera.exposure_mode = 'off'
g = camera.awb_gains #TODO: make this fixed?
camera.awb_mode = 'off'
camera.awb_gains = g
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
last_time = time.time()
image = frame.array
rawCapture.truncate(0)
if self.hsv:
hsvImage = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
hsvImage = cv2.resize(hsvImage, (len(image[0]) / self.scale_down, len(image) / self.scale_down))
green_color_mask, blue_color_mask = self.run_hsv(hsvImage, self.callback)
cv2.imshow("hsv_image", hsvImage)
else:
self.run_bgr(image, self.callback)
# cv2.imshow("rgb_image", image)
# green_color_mask = cv2.inRange(image, self.green_lower_dark_bgr, self.green_upper_dark_bgr)
# cv2.imshow("green_color_mask", green_color_mask)
#
# cv2.waitKey(1)
# stop thread
if self._stopped():
self.__simLogger.debug("Stopping camera thread")
break
# sleep if processing of camera image was faster than minimum frames per second
sleep_time = float(MIN_FPS - (time.time() - last_time))
print "sleep time: ", sleep_time
if sleep_time > 0:
time.sleep(sleep_time)
cv2.destroyAllWindows()
except Exception as e:
self.error_callback()
print "Error: ", str(e), str(sys.exc_info()[0]) + ' - ' + traceback.format_exc()
# self.__simLogger.critical("Camera exception: " + str(e) + str(
# sys.exc_info()[0]) + ' - ' + traceback.format_exc())
except (KeyboardInterrupt, SystemExit):
self.error_callback()
raise
def navigateImg():
moveTime = 0.5
sleepTime = 2
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 32
rawCapture = PiRGBArray(camera, size=(640, 480))
e1 = 20.00000
e2 = 20.00000
time.sleep(0.1)
dst = (320, 240)
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
try:
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
image = frame.array
w = 0
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
lower_blue = np.array([100,50,50])
upper_blue = np.array([120,255,255])
mask = cv2.inRange(hsv, lower_blue, upper_blue)
loc = getCenter(mask)
print(dst[1]-loc[1])
print(dst[0]-loc[0])
if(abs(dst[1]-loc[1]) > e1):
if(dst[1] > loc[1]):
print('b')
backwards(20+int((dst[1]-loc[1])/3))
else:
print('f')
forwards(20-int((dst[1]-loc[1])/3))
else:
print('fb good')
w = w + 1
if(abs(dst[0]-loc[0]) > e2):
if(dst[0] > loc[0]):
print('l')
left(20+int((dst[0]-loc[0])/3))
else:
print('r')
right(20-int((dst[0]-loc[0])/3))
else:
print('lf good')
w = w + 1
if w == 2:
print('drop')
time.sleep(moveTime)
zero()
time.sleep(sleepTime)
rawCapture.truncate(0)
except OSError:
print('io')
class VideoStream(object):
def __init__(self):
self.camera = PiCamera()
self.camera.resolution = (1920, 1080)
self.preview_window = (0, 0, 640, 480)
self.camera.framerate = 15
self.stream = None
self.rgb_array = PiRGBArray(self.camera, size=self.camera.resolution)
# initialize the frame and the variable used to indicate
# if the thread should be stopped
self.frame = None
self.preview_stopped = False
self.capture_stopped = False
def capture_hi_res(self):
self.stream = self.camera.capture_continuous(self.rgb_array, use_video_port=True, splitter_port=2, format='bgr')
Thread(target=self.capture_raw, args=()).start()
def capture_raw(self):
for f in self.stream:
self.frame = f.array
self.rgb_array.truncate(0)
if self.capture_stopped:
self.stream.close()
self.rgb_array.close()
self.camera.close()
def stop_capture(self):
self.capture_stopped = True
def preview_stream(self):
# start the thread to read frames from the video stream
Thread(target=self.preview, args=()).start()
def preview(self):
self.camera.start_preview(fullscreen=False, window=self.preview_window)
if self.preview_stopped:
self.camera.stop_preview()
def read(self):
return self.frame
def stop_preview(self):
self.preview_stopped = True
def camera(fov_v, fov_h, threshold, landing_area_x, landing_area_y):
#make the bitstream for the radio message
radio_stream = bitstream.BitStream()
# init camera and get a reference to the raw capture
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 32
rawCapture = PiRGBArray(camera, size=(640, 480))
#time for the camera to warm up which is apparently a thing, and we'd probably only have to do it once rather than every picture
time.sleep(0.1)
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
#get the data from the drone
altitude, drone_coords, heading = read_drone_data()
#get an image from the camera
edges = frame.array #TODO FROM CAMERA
rawCapture.truncate(0)
#do edge detection on the image
minVal, maxVal = find_canny_thresholds(altitude) #TODO something about these minVal, maxVal based on altitude
edges = cv2.Canny(edges, minVal, maxVal)
rows, cols = edges.shape
#create the decision grid
decision_grid, grid_cols, grid_rows, grid_pix_x, grid_pix_y = create_decision_grid(altitude, fov_h, rows, cols, landing_area_x, landing_area_y)
#create the radio message to be sent
radio_stream.write(altitude, np.uint8) #1 byte total
radio_stream.write(heading, np.uint16) #3 bytes total
radio_stream.write((grid_rows, grid_cols), np.uint16) #7 bytes total
radio_stream.write(drone_coords, float) #23 bytes total
for row in xrange(grid_rows):
for col in xrange(grid_cols):
decision_grid[row][col] = len(np.where(edges[row*grid_pix_y:row*grid_pix_y+grid_pix_y, col*grid_pix_x:col*grid_pix_x+grid_pix_x] > 0)[0])
if decision_grid[row][col] >= threshold:
radio_stream.write(1, bool)
else:
radio_stream.write(0, bool)
#radio can only handle a byte as the smallest unit of information, so we have to pad the signal
pad = 8 - ((grid_rows * grid_cols) % 8)
if pad == 8:
pad = 0
for i in xrange(pad):
radio_stream.write(0, bool)
message = radio_stream.read(str, 23 + (grid_rows*grid_cols + pad)/8)
class Camera(threading.Thread):
def __init__(self):
self.image = []
self.angle = 0
# initialize the camera and grab a reference to the raw camera capture
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.camera.framerate = 32
self.rawCapture = PiRGBArray(self.camera, size=(640, 480))
time.sleep(0.1)
#Set up thread
threading.Thread.__init__(self)
self.threadID = 100
self.name = "camera"
#Thread exits when sent to True
self.exitFlag = False
def run(self):
# capture frames from the camera
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
self.image = frame.array
# show the frame
# cv2.imshow("Frame", image)
gb_lane_cfg = LaneCfg
proc = {'houghlinesP':detect_lane_over_houghlinesP}
self.angle = proc[gb_lane_cfg['set']['proc']](gb_lane_cfg['set']['proc'], self.image, gb_lane_cfg)
#utilities.log("angle " + str(self.angle))
# clear the stream in preparation for the next frame
self.rawCapture.truncate(0)
# if the `q` key was pressed, break from the loop
if self.exitFlag == True:
break
self.camera.close()
def stop(self):
print("stoping camera")
self.exitFlag = True
def camera_loop():
use_pi = False
try:
from picamera import PiCamera
from picamera.array import PiRGBArray
use_pi = True
except OSError:
print('Couldn\'t load picamera, using OpenCV video capture')
use_pi = False
if use_pi:
print('Opening camera')
camera = PiCamera()
rawCapture = PiRGBArray(camera)
camera.resolution = (640, 480)
time.sleep(1)
camera.start_preview()
camera.awb_mode = 'off'
camera.exposure_mode = 'off'
# use initial values from config file
camera.awb_gains = (config['camera']['awb_red_gain'], config['camera']['awb_red_gain'])
camera.shutter_speed = config['camera']['shutter_speed']
camera.iso = config['camera']['iso']
print('Opened camera')
# capture frames from the camera
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image
image = frame.array
handle_image(image)
# clear the stream in preparation for the next frame
rawCapture.truncate(0)
# set all the settings if they changed in the config
camera.awb_gains = (config['camera']['awb_red_gain'], config['camera']['awb_red_gain'])
camera.shutter_speed = config['camera']['shutter_speed']
camera.iso = config['camera']['iso']
else:
print('Opening camera')
capture = cv2.VideoCapture(0)
while True:
success, img = capture.read()
if success:
handle_image(img)
class PiVideoStream(object):
def __init__(self, resolution=(320, 240), frame_rate=32):
# Init camera and Stream
self.camera = PiCamera()
self.camera.resolution = resolution
self.camera.framerate = frame_rate
self.rawCapture = PiRGBArray(self.camera, size=resolution)
self.stream = self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True)
self.frame = None
self.stopped = False
def start(self):
"""
Starts the update thread that reads the camera data
:return:
"""
Thread(target=self.update, args=()).start()
return self
def update(self):
"""
Repeatedly stores the most recently captured frame
:return:
"""
for f in self.stream:
# Store the current frame
self.frame = f.array
# Clear the array ready for the next frame
self.rawCapture.truncate(0)
if self.stopped:
self.stream.close()
self.rawCapture.close()
self.camera.close()
return
def read(self):
"""
Gets the most recently read frame
:return: Most recently captured RGB array from camera
"""
return self.frame
def stop(self):
"""
Exits out of the update function/thread
:return:
"""
self.stopped = True
class Eye:
#relays information to 4 pins where there all steppers invloving rotation should be plugged
def __init__(self, resWidth, resHeight, framerate):
self.camera = PiCamera()
self.camera.resolution = (resWidth, resHeight)
self.camera.framerate = framerate
self.rawCapture = PiRGBArray(self.camera, size=(resWidth, resHeight))
self.objectWidth = 50
self.objectHeight = 50
self.xPercent = resHeight/2;
self.yPercent = resWidth/2;
def getCenter(self):
return (self.xPercent, self.yPercent);
def recognizeFace(self, showPreview):
face_cascade = cv2.CascadeClassifier('../CAM/haar/haarcascade_frontalface_default.xml')
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
image = frame.array
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.1, 5)
for (x,y,w,h) in faces:
self.xPercent = float(x+(w/2))/float(self.camera.resolution[0]);
self.yPercent = float(y+(h/2))/float(self.camera.resolution[1]);
if showPreview:
cv2.rectangle(gray,(x,y),(x+w,y+h),(255,255,0),1)
# print the coords
print 'face percents on screen'
print (self.xPercent, self.yPercent);
# show the frame
if showPreview:
cv2.imshow("Frame", gray)
key = cv2.waitKey(1) & 0xFF
# clear the stream in preparation for the next frame
self.rawCapture.truncate(0)
# if the `q` key was pressed, break from the loop
if key == ord("c"):
break
def followFaceWithArm(self, arm):
face_cascade = cv2.CascadeClassifier('../CAM/haar/haarcascade_frontalface_default.xml')
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
image = frame.array
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.1, 5)
for (x,y,w,h) in faces:
self.xPercent = float(x+(w/2))/float(self.camera.resolution[0]);
self.yPercent = float(y+(h/2))/float(self.camera.resolution[1]);
self.objectWidth = w
self.objectHeight = h
arm.positionPercent(self.xPercent,self.yPercent)
print ('Face at percents...', self.xPercent,self.yPercent, ' Face dims...', self.objectWidth, self.objectHeight)
# go to face
key = cv2.waitKey(1) & 0xFF
# clear the stream in preparation for the next frame
self.rawCapture.truncate(0)
# if the `q` key was pressed, break from the loop
if key == ord("c"):
break
def camPreview(self):
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
image = frame.array
cv2.imshow("Frame", image)
key = cv2.waitKey(1) & 0xFF
# clear the stream in preparation for the next frame
self.rawCapture.truncate(0)
# if the `q` key was pressed, break from the loop
if key == ord("x"):
break
def videoLoop(self, outputWindow):
try:
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True):
image = frame.array
grayImg = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# cv2.putText(grayImg, 'FACE TRACKING, <f> to toggle', (0,45), 16, .35, (255,255,255), 1)
outputWindow.drawGuidesAndText(grayImg, ['MM', 'FM', 'MF'])
if outputWindow.robot.isTracking:
faces = outputWindow.robot.trackingHaar.detectMultiScale(grayImg, 1.1, 5)
for (x,y,w,h) in faces:
xPercent = float(x+(w/2))/float(outputWindow.robot.eye.camera.resolution[0]);
yPercent = float(y+(h/2))/float(outputWindow.robot.eye.camera.resolution[1]);
outputWindow.robot.arm.positionPercent(xPercent, yPercent)
grayImg= Image.fromarray(grayImg)
grayPhotoImg = ImageTk.PhotoImage(grayImg)
if outputWindow.panel is None: # create and mount panel if it's not there
outputWindow.panel = Label(outputWindow.frame, image=grayPhotoImg)
outputWindow.panel.image = grayPhotoImg
outputWindow.panel.pack(padx=10, pady=10)
else:
outputWindow.panel.configure(image=grayPhotoImg)
outputWindow.panel.image = grayPhotoImg
self.rawCapture.truncate(0)
except RuntimeError, e:
print("[INFO] caught a RuntimeError")
# stream = picamera.PiCameraCircularIO(camera, seconds = video_preseconds)
stream = picamera.PiCameraCircularIO(camera, size=3000000)
# video recording into circular buffer from splitter port 1
camera.start_recording(stream, format='h264', splitter_port=1)
#camera.start_recording('test.h264', splitter_port=1)
# wait 2 seconds for stable video data
camera.wait_recording(2, splitter_port=1)
# motion_detected = False
print(camera.resolution)
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text???????????????????
rawCapture.truncate()
rawCapture.seek(0)
frame2 = frame.array
frameNo = frameNo + 1
img_rgb = frame2
frame2arm = getCroppedImage(frame2, resetArmCrops)
img_gray2arm = cv2.cvtColor(frame2arm, cv2.COLOR_BGR2GRAY)
# tripSense()
# print("GPIO", GPIO.input(sensor[0]))
# print('Sensore', GPIO.input(sensor[1]), GPIO.input(sensor[2]))
while (GPIO.input(sensor[0]) == GPIO.HIGH):
# GPIO.output((segment7All[ballCounter % 10]), GPIO.LOW)
# print('Ball Timer Awake ', ballCounter)
done = False
def capture_thread(self, IPinver):
global frame_image, camera
ap = argparse.ArgumentParser() # OpenCV initialization
ap.add_argument("-b",
"--buffer",
type=int,
default=64,
help="max buffer size")
args = vars(ap.parse_args())
pts = deque(maxlen=args["buffer"])
font = cv2.FONT_HERSHEY_SIMPLEX
camera = picamera.PiCamera()
camera.resolution = (640, 480)
camera.framerate = 20
rawCapture = PiRGBArray(camera, size=(640, 480))
context = zmq.Context()
footage_socket = context.socket(zmq.PUB)
print(IPinver)
footage_socket.connect("tcp://%s:5555" % IPinver)
avg = None
motionCounter = 0
lastMovtionCaptured = datetime.datetime.now()
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
frame_image = frame.array
timestamp = datetime.datetime.now()
if FindColorMode:
####>>>OpenCV Start<<<####
hsv = cv2.cvtColor(frame_image, cv2.COLOR_BGR2HSV)
mask = cv2.inRange(hsv, colorLower, colorUpper) # 1
mask = cv2.erode(mask, None, iterations=2)
mask = cv2.dilate(mask, None, iterations=2)
cnts = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)[-2]
center = None
if len(cnts) > 0:
cv2.putText(
frame_image,
"Target Detected",
(40, 60),
font,
0.5,
(255, 255, 255),
1,
cv2.LINE_AA,
)
c = max(cnts, key=cv2.contourArea)
((x, y), radius) = cv2.minEnclosingCircle(c)
M = cv2.moments(c)
center = (int(M["m10"] / M["m00"]),
int(M["m01"] / M["m00"]))
X = int(x)
Y = int(y)
if radius > 10:
cv2.rectangle(
frame_image,
(int(x - radius), int(y + radius)),
(int(x + radius), int(y - radius)),
(255, 255, 255),
1,
)
if Y < (240 - tor):
error = (240 - Y) / 5
outv = int(round((pid.GenOut(error)), 0))
servo.up(outv)
Y_lock = 0
elif Y > (240 + tor):
error = (Y - 240) / 5
outv = int(round((pid.GenOut(error)), 0))
servo.down(outv)
Y_lock = 0
else:
Y_lock = 1
if X < (320 - tor * 3):
error = (320 - X) / 5
outv = int(round((pid.GenOut(error)), 0))
servo.lookleft(outv)
servo.turnLeft(coe_Genout(error, 64))
X_lock = 0
elif X > (330 + tor * 3):
error = (X - 240) / 5
outv = int(round((pid.GenOut(error)), 0))
servo.lookright(outv)
servo.turnRight(coe_Genout(error, 64))
X_lock = 0
else:
move.motorStop()
X_lock = 1
if X_lock == 1 and Y_lock == 1:
switch.switch(1, 1)
switch.switch(2, 1)
switch.switch(3, 1)
moveCtrl(ultra.checkdist(), back_R, forward_R)
else:
move.motorStop()
switch.switch(1, 0)
switch.switch(2, 0)
switch.switch(3, 0)
else:
cv2.putText(
frame_image,
"Target Detecting",
(40, 60),
font,
0.5,
(255, 255, 255),
1,
cv2.LINE_AA,
)
move.motorStop()
####>>>OpenCV Ends<<<####
if WatchDogMode:
gray = cv2.cvtColor(frame_image, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (21, 21), 0)
if avg is None:
print("[INFO] starting background model...")
avg = gray.copy().astype("float")
rawCapture.truncate(0)
continue
cv2.accumulateWeighted(gray, avg, 0.5)
frameDelta = cv2.absdiff(gray, cv2.convertScaleAbs(avg))
# threshold the delta image, dilate the thresholded image to fill
# in holes, then find contours on thresholded image
thresh = cv2.threshold(frameDelta, 5, 255,
cv2.THRESH_BINARY)[1]
thresh = cv2.dilate(thresh, None, iterations=2)
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
# print('x')
# loop over the contours
for c in cnts:
# if the contour is too small, ignore it
if cv2.contourArea(c) < 5000:
continue
# compute the bounding box for the contour, draw it on the frame,
# and update the text
(x, y, w, h) = cv2.boundingRect(c)
cv2.rectangle(frame_image, (x, y), (x + w, y + h),
(128, 255, 0), 1)
text = "Occupied"
motionCounter += 1
# print(motionCounter)
# print(text)
LED.colorWipe(255, 16, 0)
lastMovtionCaptured = timestamp
switch.switch(1, 1)
switch.switch(2, 1)
switch.switch(3, 1)
if (timestamp - lastMovtionCaptured).seconds >= 0.5:
LED.colorWipe(255, 255, 0)
switch.switch(1, 0)
switch.switch(2, 0)
switch.switch(3, 0)
if FindLineMode:
cvFindLine()
cv2.line(frame_image, (300, 240), (340, 240), (128, 255, 128), 1)
cv2.line(frame_image, (320, 220), (320, 260), (128, 255, 128), 1)
if FindLineMode and not frameRender:
encoded, buffer = cv2.imencode(".jpg", frame_findline)
else:
encoded, buffer = cv2.imencode(".jpg", frame_image)
jpg_as_text = base64.b64encode(buffer)
footage_socket.send(jpg_as_text)
rawCapture.truncate(0)
class Camera(HDThread):
def __init__(self, thread_name, logging, detection_queue, visibility_queue,
target_fps):
super().__init__(thread_name, logging, target_fps)
self.detection_queue = detection_queue # type: queue.Queue
self.visibility_queue = visibility_queue
try:
self.logging.info("{} - Init PiCamera...".format(thread_name))
from picamera import PiCamera
from picamera.array import PiRGBArray
self.camera = PiCamera()
self.camera.resolution = (640, 480)
self.rawCapture = PiRGBArray(self.camera)
self.picamera_mode = True
self.logging.info("{} - Init PiCamera success".format(thread_name))
except:
self.logging.info("{} - Init VideoCapture...".format(thread_name))
self.camera = cv2.VideoCapture(0)
self.logging.info(
"{} - Init VideoCapture success".format(thread_name))
self.picamera_mode = False
def _run(self) -> None:
image = self._from_camera()
# wait until queue is empty
if self.detection_queue.full():
self.detection_queue.get()
self.detection_queue.put(image)
if self.visibility_queue.full():
self.visibility_queue.get()
self.visibility_queue.put(image)
def is_module_in_error(self):
return self.in_error
def _from_camera(self):
temp_time = start_time = datetime.now()
self.logging.debug("{} - Start.".format(self.thread_name))
if self.picamera_mode:
self.camera.capture(self.rawCapture,
format="bgr",
use_video_port=True)
img = self.rawCapture.array
else:
ret, img = self.camera.read()
self.logging.debug("{} - Capturing image. Duration={}".format(
self.thread_name,
datetime.now() - temp_time))
if self.picamera_mode:
temp_time = datetime.now()
self.rawCapture.truncate(0)
self.logging.debug("{} - Capturing truncate Duration={}".format(
self.thread_name,
datetime.now() - temp_time))
iteration_time = datetime.now() - start_time
self.iteration_time_sec = iteration_time.microseconds / 1000000
self.logging.debug("{} - End. Total Duration={}".format(
self.thread_name, iteration_time))
return img
def main():
# create logger'
logger = logging.getLogger('home_security')
logger.setLevel(logging.DEBUG)
# create file handler which logs even debug messages
fh = logging.FileHandler('home_security.log')
fh.setLevel(logging.DEBUG)
# create console handler with a higher log level
ch = logging.StreamHandler()
ch.setLevel(logging.INFO)
# create formatter and add it to the handlers
formatter = logging.Formatter(
'%(asctime)s - %(name)s - %(levelname)s - %(message)s')
fh.setFormatter(formatter)
ch.setFormatter(formatter)
# add the handlers to the logger
logger.addHandler(fh)
logger.addHandler(ch)
#syslog = logging.handlers.SysLogHandler(address = '/dev/log')
#syslog.setLevel(logging.ERROR)
#logger.addHandler(syslog)
# construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-c",
"--conf",
required=True,
help="path to the JSON configuration file")
args = vars(ap.parse_args())
# filter warnings, load the configuration and initialize the Dropbox
# client
warnings.filterwarnings("ignore")
conf = json.load(open(args["conf"]))
client = None
# check to see if the Dropbox should be used
if conf["use_dropbox"]:
if conf["accessToken"]:
accessToken = conf["accessToken"]
userID = "*****@*****.**"
else:
# connect to dropbox and start the session authorization process
#flow = DropboxOAuth2FlowNoRedirect(conf["dropbox_key"], conf["dropbox_secret"])
#print "[INFO] Authorize this application: {}".format(flow.start())
#authCode = raw_input("Enter auth code here: ").strip()
# finish the authorization and grab the Dropbox client
#(accessToken, userID) = flow.finish(authCode)
print " ************* error *************"
print "accessToken:{} userID:{}".format(accessToken, userID)
# Create a dropbox object using an API v2 key
dbx = dropbox.Dropbox(token)
#client = DropboxClient(accessToken)
print "[SUCCESS] dropbox account linked"
# initialize the camera and grab a reference to the raw camera capture
camera = PiCamera()
camera.resolution = tuple(conf["resolution"])
camera.framerate = conf["fps"]
rawCapture = PiRGBArray(camera, size=tuple(conf["resolution"]))
# allow the camera to warmup, then initialize the average frame, last
# uploaded timestamp, and frame motion counter
print "[INFO] warming up..."
time.sleep(conf["camera_warmup_time"])
avg = None
lastUploaded = datetime.datetime.now()
dayNumber = lastUploaded.toordinal()
motionCounter = 0
# capture frames from the camera
for f in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
# grab the raw NumPy array representing the image and initialize
# the timestamp and movement flag
frame = f.array
timestamp = datetime.datetime.now()
dayNumberNow = timestamp.toordinal()
movement = False
# resize the frame, convert it to grayscale, and blur it
frame = imutils.resize(frame, width=600)
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (21, 21), 0)
# if the average frame is None, initialize it
if avg is None:
print "[INFO] starting background model..."
avg = gray.copy().astype("float")
rawCapture.truncate(0)
continue
# accumulate the weighted average between the current frame and
# previous frames, then compute the difference between the current
# frame and running average
cv2.accumulateWeighted(gray, avg, 0.5)
frameDelta = cv2.absdiff(gray, cv2.convertScaleAbs(avg))
# threshold the delta image, dilate the thresholded image to fill
# in holes, then find contours on thresholded image
thresh = cv2.threshold(frameDelta, conf["delta_thresh"], 255,
cv2.THRESH_BINARY)[1]
thresh = cv2.dilate(thresh, None, iterations=2)
(_, cnts, _) = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
# loop over the contours. 0,0 is tlc. y increases down, x increase right
x, y = 0, 0
for c in cnts:
(x, y, w, h) = cv2.boundingRect(c)
# if the contour is too small, y co-ord is too low ignore it
if (cv2.contourArea(c) < conf["min_area"]) or ((y + h) < 320):
continue
# compute the bounding box for the contour, draw it on the frame,
# and update the text
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
movement = True
# draw the text and timestamp on the frame
ts = timestamp.strftime("%A %d %B %Y %I:%M:%S%p")
cv2.putText(frame, "x: {} y: {}".format(x, y), (10, 20),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 0), 2)
cv2.putText(frame, ts, (10, frame.shape[0] - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.35, (255, 255, 255), 1)
# check to see if there is movement
if movement:
logger.info("movement detected")
# check to see if enough time has passed between uploads
if (timestamp -
lastUploaded).seconds >= conf["min_upload_seconds"]:
# increment the motion counter
motionCounter += 1
# check to see if the number of frames with consistent motion is
# high enough
if motionCounter >= conf["min_motion_frames"]:
# check to see if dropbox should be used
if conf["use_dropbox"]:
# write the image to temporary file
t = TempImage()
cv2.imwrite(t.path, frame)
suffix = (dayNumberNow % 20) + 1 #(1..20)
new_path = "Public/SecurityDawson65_" + str(suffix)
# upload the image to Dropbox and cleanup the tempory image
try:
path = "{base_path}/{timestamp}.jpg".format(
base_path=new_path, timestamp=ts)
logger.info("[UPLOAD] {}".format(path))
#client.put_file(path, open(t.path, "rb"))
# we want to overwite any previous version of the file
contents = open(t.path, "rb").read()
meta = dbx.files_upload(
contents,
path,
mode=dropbox.files.WriteMode("overwrite"))
except Exception as e:
logger.exception("Network error. Upload failed")
time.sleep(30) #wait for dropbox to recover
finally:
t.cleanup()
# update the last uploaded timestamp and reset the motion
# counter
lastUploaded = timestamp
motionCounter = 0
else:
logger.info(
"failed min_motion_frames {}".format(motionCounter))
else:
logger.info("failed min_upload_seconds")
# otherwise, no movement detected
else:
motionCounter = 0
if dayNumber != dayNumberNow:
#midnight. clear new folder
suffix = (dayNumberNow % 20) + 1 #(1..20)
new_path = "Public/SecurityDawson65_" + str(suffix)
delete_files(dbx, logger, new_path)
dayNumber = dayNumberNow
logger.info("old files deleted for day %s" %
str(dayNumberNow % 20 + 1))
# check to see if the frames should be displayed to screen
if conf["show_video"]:
# display the security feed
cv2.imshow("Security Feed", frame)
key = cv2.waitKey(1) & 0xFF
# if the `q` key is pressed, break from the loop
if key == ord("q"):
break
# clear the stream in preparation for the next frame
rawCapture.truncate(0)
class VideoCapturer:
multiplier = int()
height = int()
width = int()
camera = object()
filename = str()
rawCapture = object()
fifoFilters = []
snapRate = 20
def __init__(self, _height, _width, mult, frame_rate):
self.multiplier = mult
self.height = int(mult * _height)
self.width = int(mult * _width)
self.camera = PiCamera()
self.camera.resolution = (self.height, self.width)
self.camera.framerate = frame_rate
self.rawCapture = PiRGBArray(self.camera,
size=(self.height, self.width))
time.sleep(0.1)
self.filename = "temp.mp4"
def runCam(self, loadFilters=False):
count = 0
prev = float()
for frame in self.camera.capture_continuous(self.rawCapture,
format="bgr",
use_video_port=True):
image = frame.array
image = self.rotateImage(image, 270)
b, g, r = cv2.split(image)
if loadFilters:
for funct in self.fifoFilters:
b = funct(b)
else:
b = image
count = count + 1
if count > 0:
b = 0.5 * (b + prev)
prev = b
cv2.imshow("Frame", b)
if (count > 0 and count % self.snapRate == 0):
cv2.imwrite(str(count) + ".png", b)
key = cv2.waitKey(1) & 0xFF
self.rawCapture.truncate(0)
if key == "ord":
break
self.camera.close()
cv2.destroyAllWindows()
# works for single channel only
def addFilter(self, newFilterFunction):
self.fifoFilters.append(newFilterFunction)
def rotateImage(self, arr, angle):
if angle > 89:
arr = np.rot90(arr)
if angle > 179:
arr = np.rot90(arr)
if angle > 269:
arr = np.rot90(arr)
return arr
def cam(self):
# initialize the camera and grab a reference to the raw camera capture
camera = PiCamera()
height = 500
width = 500
camera.resolution = (width, height)
camera.framerate = 60
#camera.color_effects=(128,128)
camera.iso = 500
#camera.image_effect='denoise'
#camera.exposure_mode = 'night'
rawCapture = PiRGBArray(camera, size=(width, height))
# allow the camera to warmup
time.sleep(0.5)
logging.basicConfig(
level=logging.DEBUG,
format='[%(levelname)s] (%(threadName)-9s) %(message)s',
)
try:
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
# grab the raw NumPy array representing the image
image = frame.array
logging.debug("Thread")
#print image
#print cv2.mean(image)
template_y = 0
template_x = 0
grid = 0
width_size = int(width / self.grid_column)
height_size = int(height / self.grid_row)
for i in range(self.grid_row):
template_x = 0
for j in range(self.grid_column):
#t = threading.Thread(target=process_eachgrid, args=(image,template_x,template_y,grid,width_size,height_size))
#t.start()
#t.join()
self.process_eachgrid(image, template_x, template_y,
grid, width_size, height_size)
template_x = template_x + width_size
grid = grid + 1
template_y = template_y + height_size
self.draw_grid(image, height, width, width_size, height_size)
#print com.motion_start
# clear the stream in preparation for the next frame
rawCapture.truncate(0)
# show the image
cv2.imshow("Security Feed", image)
key = cv2.waitKey(1) & 0xFF
# if the `q` key was pressed, break from the loop
if key == ord("q"):
break
#if not com.motion_start:
#break
finally:
#com.motion_start= False
cv2.destroyWindow("Security Feed")
import time
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 32
rawCapture = PiRGBArray(camera, size=(640, 480))
cascPath = sys.argv[0]
faceCascade = cv2.CascadeClassifier("haarcascade_frontalface_default.xml")
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
# Capture frame-by-frame
frame = frame.array
rawCapture.truncate(0)
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = faceCascade.detectMultiScale(
gray,
scaleFactor=1.1,
minNeighbors=5,
minSize=(30, 30),
)
# Draw a rectangle around the faces
for (x, y, w, h) in faces:
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
# Display the resulting frame
cv2.imshow('Video', frame)
def run():
relay_pin = [26]
GPIO.setmode(GPIO.BCM)
GPIO.setup(relay_pin, GPIO.OUT)
GPIO.output(relay_pin, 0)
with open('labels', 'rb') as f:
dict = pickle.load(f)
f.close()
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 30
raw_capture = PiRGBArray(camera, size=(640, 480))
face_cascade = cv2.CascadeClassifier("haarcascade_face.xml")
recognizer = cv2.face.LBPHFaceRecognizer_create()
recognizer.read("trainer.yml")
font = cv2.FONT_HERSHEY_SIMPLEX
init_time = 0
playback_started = False
for frame in camera.capture_continuous(raw_capture,
format="bgr",
use_video_port=True):
frame = frame.array
# flips the camera
frame = cv2.flip(frame, -1)
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray,
scaleFactor=1.5,
minNeighbors=5)
for (x, y, w, h) in faces:
roi_gray = gray[y:y + h, x:x + w]
id_, conf = recognizer.predict(roi_gray)
highest_conf = 0
for name, value in dict.items():
if value == id_:
print(name + " conf: " + str(conf))
highest_conf = name
if conf >= 70 and conf < 105:
GPIO.output(relay_pin, 1)
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
cv2.putText(frame, highest_conf + str(conf), (x, y), font, 2,
(0, 0, 255), 2, cv2.LINE_AA)
# music section
with open("user_profiles.json", "r") as user_profiles:
user_dict = json.load(user_profiles)
try:
if (not playback_started):
song = user_dict[highest_conf]
pygame.mixer.init()
pygame.mixer.music.load(song)
pygame.mixer.music.play()
init_time = time.time()
playback_started = True
if (playback_started):
curr_time = time.time()
if curr_time - init_time > 10:
playback_started = False
pygame.mixer.music.fadeout(5000)
#while pygame.mixer.music.get_busy() == True:
# pygame.time.Clock().tick(100000)
# break;
# pygame.mixer.music.fadeout(20000)
except KeyError:
print(
"Song not found! Please set up your user profile.")
else:
GPIO.output(relay_pin, 0)
cv2.imshow('frame', frame)
key = cv2.waitKey(1)
raw_capture.truncate(0)
if key == 27:
break
cv2.destroyAllWindows()
class VideoStream(object):
def __init__(self, src=None, usePiCamera=False, resolution=(320, 240),
framerate=30, format='bgr', trigger=None):
# initialize the camera and stream
if usePiCamera:
self.camera = PiCamera()
else:
if isinstance(src, (UnifiedCamera, PiCamera)):
self.camera = src
else:
self.camera = UnifiedCamera(src)
self.resolution = resolution
self.framerate = framerate
self.rawCapture = PiRGBArray(self.camera, size=resolution)
self._warm_up_time = 2
self._creation_time = time()
# initialize the frame and the variable used to indicate
# if the thread should be stopped
self.frame = None
self._stop = True # thread is stopped or non existent
if trigger is None:
trigger = Event()
self.trigger = trigger
self.format = format
self._thread_free = Event()
self._order = Event()
self._lock = RLock()
self._thread = None
self._frame_cache = None
# the optimization flag lets the camera wait until it is needed
self._optimize = True
@property
def resolution(self):
return self.camera.resolution
@resolution.setter
def resolution(self, value):
self.camera.resolution = value
@property
def framerate(self):
return self.camera.framerate
@framerate.setter
def framerate(self, value):
self.camera.framerate = value
def _update_func(self):
#print("camera thread {} started".format(self._thread))
# initialize events
#toggle = True
try:
df = time() - self._creation_time
if df < self._warm_up_time:
# allow to warm up if camera is opened too quickly
sleep(self._warm_up_time-df)
self.camera.start_preview() # start camera
self.frame = self.camera.capture(self.rawCapture, self.format)
if self.frame is None:
raise CameraError("camera '{}' not working".format(self.camera))
self._thread_free.set() # notify thread is free to receive orders
# keep looping infinitely until the thread is stopped
while not self._stop:
# because self.trigger can be shared it is possible that
# self.read function was called setting self.trigger once
# but this can be turned off while in other threads so
# we have to check self.thread_free is not set meaning
# that it has to produce a frame, thus it should not
# be blocked
if self._optimize:
# self.trigger.clear() # make the thread wait until event
# #self.__debug("event waiting in {}'s thread".format(id(self)))
if self.framerate:
self.trigger.wait(1/self.framerate) # wait until read function or event calls
else:
self.trigger.wait(10)
# self._thread_free.clear() # tell thread is busy
# #self.__debug("event was unblocked in {}'s thread".format(id(self)))
# self._order.set() # there is an order
# if the thread indicator variable is set, stop the thread
# and resource camera resources
#if self._stop:
# break
self.frame = self.camera.capture(self.rawCapture, self.format)
#print("{}taking photo in {}".format((""," ")[toggle],id(self)))
#toggle = not toggle
# notify frame was produced and thread is free
# as quickly as possible
#self._thread_free.set()
self.rawCapture.truncate(0)
if self._frame_cache is None and self.trigger.is_set():
self._frame_cache = self.frame
self._order.set()
finally:
# ending thread
#self.rawCapture.close()
self.camera.close()
self._stop = True
self._thread_free.set() # prevents blocking in main
self._order.set() # prent blocking in main
#print("camera thread {} ended".format(self._thread))
def read(self):
if self.trigger.is_set():
self._order.wait()
return self._frame_cache
return self.frame
def clear_order(self):
self._frame_cache = None
self._order.clear()
def get_frame(self):
"""
safely give frame from latest read
"""
if self._stop:
raise RuntimeError("{} must be started".format(type(self)))
return self.read()
def start(self):
# start the thread to read frames from the video stream
with self._lock:
# if several threads are trying to start it wait
# if while this lock was waiting and this thread ended in another
# lock, open the tread again to prevent inconsistencies
if self.closed():
self._thread = t = Thread(target=self._update_func, args=())
t.daemon = False
self._stop = False # thread started
self._thread_free.clear()
t.start()
self._thread_free.wait(10)
if self._stop:
raise CameraError("camera '{}' not ready".format(self.camera))
return self
def close(self):
with self._lock:
if not self.closed():
# indicate that the thread should be stopped
self._stop = True
self._thread_free.clear()
self.trigger.set() # un-pause threads
self._thread.join(10)
if not self.closed():
raise Exception("Thread didn't close")
def closed(self):
with self._lock:
return self._thread is None or not self._thread.is_alive()
def __enter__(self):
self.start()
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.close()
@classmethod
def __debug(cls, data=None, start_debug=False):
if not hasattr(cls, "checks"):
cls.checks = Counter()
cls.check_cum = set()
if start_debug and data in cls.check_cum:
cls.checks[frozenset(cls.check_cum)] += 1
cls.check_cum = set()
cls.check_cum.add(data)
else:
cls.check_cum.add(data)
class Camera(MultiProcessing):
_showImage = True
_camera = None
_rawCapture = None
_cascade = None
_nested = None
__cameraResolutionX = 640
__cameraResolutionY = 480
__posXFaceKey = MultiProcessing.get_next_key(
) #-1 # -1=no face, 0=max left, 1=max right
__posYFaceKey = MultiProcessing.get_next_key(
) #-1 # -1=no face, 0=max bottom, 1=max top
_released = False
_delay_seconds = 1
_delay_seconds_when_idle = 1.5
def __init__(self):
print("camera init")
super().__init__(prio=20)
self.posXFace = -1
self.posYFace = -1
super().StartUpdating()
## multi process properties START ##
@property
def posXFace(self):
return self.GetSharedValue(self.__posXFaceKey)
@posXFace.setter
def posXFace(self, value):
self.SetSharedValue(self.__posXFaceKey, value)
@property
def posYFace(self):
return self.GetSharedValue(self.__posYFaceKey)
@posYFace.setter
def posYFace(self, value):
self.SetSharedValue(self.__posYFaceKey, value)
## multi process properties END ##
def detect(self, img, cascade):
rects = cascade.detectMultiScale(
img,
scaleFactor=1.3,
minNeighbors=3,
minSize=(int(self.__cameraResolutionX / 15),
int(self.__cameraResolutionY / 15)),
flags=cv2.CASCADE_SCALE_IMAGE)
if len(rects) == 0:
return []
rects[:, 2:] += rects[:, :2]
return rects
def draw_rects(self, img, rects, color):
for x1, y1, x2, y2 in rects:
cv2.rectangle(img, (x1, y1), (x2, y2), color, 2)
def Update(self):
print("camera start")
cv2.destroyAllWindows()
# initialize the camera and grab a reference to the raw camera capture
self._camera = PiCamera()
self._camera.resolution = (self.__cameraResolutionX,
self.__cameraResolutionY)
self._camera.framerate = 32
self._rawCapture = PiRGBArray(
self._camera
) #, size=(self._camera.resolution.width, self._camera.resolution.height))
# allow the camera to warmup
time.sleep(0.1)
cascade_fn = "/home/pi/opencv-3.1.0/data/haarcascades/haarcascade_frontalface_alt.xml"
nested_fn = "/home/pi/opencv-3.1.0/data/haarcascades/haarcascade_eye.xml"
#cascade_fn = args.get('--cascade', "../../data/haarcascades/haarcascade_frontalface_default.xml")
#nested_fn = args.get('--nested-cascade', "../../data/haarcascades/haarcascade_eye.xml")
self._cascade = cv2.CascadeClassifier(cascade_fn)
self._nested = cv2.CascadeClassifier(nested_fn)
for frame in self._camera.capture_continuous(self._rawCapture,
format="bgr",
use_video_port=True):
if (super().updating_ended == True):
return
# grab the raw NumPy array representing the image, then initialize the timestamp
# and occupied/unoccupied text
image = frame.array
# local modules
#from video import create_capture
from common import clock, draw_str
#self._camera.capture(self._rawCapture, format="bgr")
#image = self._rawCapture.array
cv2.imshow('image', image)
# clear the stream in preparation for the next frame
self._rawCapture.truncate(0)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
gray = cv2.equalizeHist(gray)
t = clock()
rects = self.detect(gray, self._cascade)
if (self._showImage == True):
vis = image.copy()
self.draw_rects(vis, rects, (0, 255, 0))
dt = 0
found_something = False
if not self._nested.empty():
posX = -1
posY = -1
bestWidth = -1
for x1, y1, x2, y2 in rects:
width = x2 - x1
if (width > bestWidth):
bestWidth = width
posX = (x1 +
(x2 - x1) / 2) / self._camera.resolution.width
posY = (y1 +
(y2 - y1) / 2) / self._camera.resolution.height
if (self._showImage == True):
roi = gray[y1:y2, x1:x2]
vis_roi = vis[y1:y2, x1:x2]
subrects = self.detect(roi.copy(), self._nested)
self.draw_rects(vis_roi, subrects, (255, 0, 0))
self.posXFace = posX
self.posYFace = posY
dt = clock() - t
if (posX != -1):
#print('camera time: %.1f ms' % (dt*1000))
found_something = True
if (self._showImage == True):
draw_str(vis, (20, 20), 'time: %.1f ms' % (dt * 1000))
cv2.imshow('facedetect', vis)
if (found_something == True):
time.sleep(self._delay_seconds)
else:
time.sleep(self._delay_seconds_when_idle)
def ResetFace(self):
self.posXFace = -1
self.posYFace = -1
def Release(self):
if (self._released == False):
self._released = True
print("shutting down camera")
super().EndUpdating()
def __del__(self):
self.Release()
font = cv2.FONT_HERSHEY_SIMPLEX
fps_navg = 5 # how many frames to average over for the FPS timer
fps_times = []
fourcc = cv2.VideoWriter_fourcc(*"MJPG")
out = cv2.VideoWriter("output2.avi", fourcc, 5.0, (447, 448))
with picam as camera:
camera.resolution = (xres, yres)
stream = PiRGBArray(camera, size=(xres, yres))
for frame in camera.capture_continuous(stream,
format="bgr",
use_video_port=True):
image = frame.array
image = cv2.flip(image, -1)
#cv2.circle(image, (448, 416), 100, (0, 0, 0), -1)
image = image[int(yres / 4.8):yres - int(yres / 3.318),
int(xres / 4.053):xres - int(xres / 3.8)].copy()
cv2.imshow("Press q to exit", image)
out.write(image)
key = cv2.waitKey(1) & 0xFF
stream.truncate(0)
if key == ord("q"):
out.release()
break
right.stop()
else:
#move forward
left.start(100)
right.start(100)
sleep(0)
forward()
sleep(0.2)
left.stop()
right.stop()
if cv2.waitKey(1) == 27:
break
##################################################################################################################
raw_cap.truncate(0)
#raw_cap.seek(0)
frame_cnt = frame_cnt + 1
#if the picam has captured 10 seconds of video leave the loop and stop recording
if (frame_cnt > 100):
#cam.stop_preview()
#cam.close()
break
###################################################################################################################
print "Ending...."
GPIO.cleanup()
cv2.destroyAllWindows()
class camera():
def __init__(self):
conf_file = open(
"/home/max/projects/catflap/ros_catkin_ws/src/catflap_image_collector/scripts/conf.yaml",
'r')
self.cfg = yaml.load(conf_file)
conf_file.close()
self.telegram_publisher = rospy.Publisher('telegram_message',
String,
queue_size=150)
def camera_init(self):
self.camera = PiCamera()
self.camera.resolution = self.cfg["camera"]["resolution"]
self.camera.rotation = self.cfg["camera"]["rotation"]
self.camera.framerate = self.cfg["camera"]["framerate"]
self.rawCapture = PiRGBArray(self.camera,
size=self.cfg["camera"]["resolution"])
self.telegram_publisher.publish("camera init")
self.lock = Lock()
def classifier_init(self):
self.cascade = cv2.CascadeClassifier(
self.cfg["camera"]["classifier"]["path"])
bg_temp = cv2.imread(self.cfg["camera"]["background_image"])
bg_temp = bg_temp[:, 150:390, :]
self.bg = cv2.cvtColor(bg_temp, cv2.COLOR_BGR2GRAY)
def get_cat_faces(self):
pass
def detect_prey(self, cnt, h, area_threshold=200):
# check if a contour reaches below height h
# if the contour are below height h is above the threshold, return True
prey_detected = False
ind_list = []
for i, point in enumerate(cnt):
if cnt[i][0][1] <= h:
ind_list.append(i)
prey_cnt = np.delete(cnt, ind_list, 0)
if len(prey_cnt) > 0 and cv2.contourArea(prey_cnt) >= area_threshold:
prey_detected = True
else:
prey_detected = False
return (prey_detected, prey_cnt)
def move_cnts(self, cnt, x, y):
for i, point in enumerate(cnt):
cnt[i][0][0] = cnt[i][0][0] + x
cnt[i][0][1] = cnt[i][0][1] + y
return cnt
def get_biggest_contour(self, cnts):
areas = []
for c in cnts:
areas.append(cv2.contourArea(c))
sortedCnts = sorted(zip(areas, cnts), key=lambda x: x[0], reverse=True)
return sortedCnts[0][1]
def action(self, trust):
self.lock.acquire()
# take a picture, detect
prey_detected = False
catsnout_detected = False
counter = 6
settings = ""
for frame in self.camera.capture_continuous(self.rawCapture,
format="bgr",
use_video_port=True):
self.image = frame.array
self.rawCapture.truncate(0)
counter -= 1
if counter < 1:
break
n = datetime.now()
timestamp_text = "{0:04d}_{1:02d}_{2:02d}_{3:02d}_{4:02d}_{5:02d}_{6:01d}".format(
n.year, n.month, n.day, n.hour, n.minute, n.second,
int(n.microsecond / 100000))
settings = "{0}: ".format(timestamp_text)
settings = settings + "{0}, {1}, {2}, {3}, {4}, {5}, {6}, {7}, {8}, {9}, {10}, {11}, {12}, {13}, {14}, {15}, {16}, {17}".format(
self.camera.sharpness, self.camera.contrast,
self.camera.brightness, self.camera.saturation, self.camera.ISO,
self.camera.exposure_compensation, self.camera.exposure_mode,
self.camera.meter_mode, self.camera.awb_mode,
self.camera.analog_gain, self.camera.awb_gains,
self.camera.digital_gain, self.camera.drc_strength,
self.camera.exposure_speed, self.camera.iso,
self.camera.saturation, self.camera.sensor_mode,
self.camera.shutter_speed)
settings = settings + "\n"
try:
file = open("/home/max/camera_settings.txt", "a")
file.write(settings)
file.close()
except Exception:
rospy.logdebug("Unexpected Error: {0}".format(sys.exc_info()[0]))
filename = "/home/max/Pictures/raw/{0}.png".format(timestamp_text)
cv2.imwrite(filename, self.image)
# region of interest, background area
self.image = self.image[:, 150:390, :]
image_text = "no catsnout detected"
## detection
rects = self.cascade.detectMultiScale(self.image,
scaleFactor=1.1,
minNeighbors=5,
minSize=(67, 50),
maxSize=(240, 195),
flags=cv2.CASCADE_SCALE_IMAGE)
if rects != ():
catsnout_detected = True
# reset picture sequence when catsnout was detected
self.picture_sequence = 0
for (x, y, w, h) in rects:
(x1, y1) = (x + int(w * 4. / 5.), y + h + h)
if y1 > self.image.shape[0]:
y1 = self.image.shape[0]
# paint catsnout detection window in green
cv2.rectangle(self.image, (x, y), (x1, y1), (0, 255, 0), 2)
# grayscale image
img = cv2.cvtColor(self.image, cv2.COLOR_BGR2GRAY)
# histogram equalization
img = cv2.equalizeHist(img)
# extract region of interes
roi = img[y:y1, x:x1]
# extract region of interest in the background image
bg_roi = self.bg[y:y1, x:x1]
bg_roi = cv2.multiply(bg_roi, .6)
# subtract the roi from background
diff_roi = cv2.subtract(roi, bg_roi)
blur = cv2.GaussianBlur(diff_roi, (5, 5), 0)
ret, thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_OTSU)
# erode and dilate -> remove small blobs
#thresh = cv2.erode(thresh, None, iterations=1)
#thresh = cv2.dilate(thresh, None, iterations=2)
#thresh = cv2.erode(thresh, None, iterations=1)
#thresh = cv2.dilate(thresh, None, iterations=2)
# invert
thresh = cv2.bitwise_not(thresh)
# find contours
image_c, cnts, hierarchy = cv2.findContours(thresh, 1, 2)
# look for biggest contour
biggest_cnt = self.get_biggest_contour(cnts)
# shift contour coordinates from roi to orig image
biggest_cnt_moved = self.move_cnts(biggest_cnt, x, y)
# draw contour
cv2.drawContours(self.image, [biggest_cnt_moved],
-1, (0, 60, 100),
thickness=1)
# detect prey
prey_detected, prey_cnt = self.detect_prey(biggest_cnt_moved,
y + int(h / 5 * 4))
if prey_detected:
# draw detection in red
cv2.drawContours(self.image, [prey_cnt],
-1, (0, 0, 255),
thickness=2)
image_text = "prey detected"
trust = trust * 0.33
filename_mod = "/home/max/Pictures/classified/cat_prey/{0}_prey.png".format(
timestamp_text)
else:
image_text = "no prey detected"
trust = trust * 2.0
filename_mod = "/home/max/Pictures/classified/cat_no_prey/{0}_no_prey.png".format(
timestamp_text)
else:
trust = trust * 0.9
filename_mod = "/home/max/Pictures/classified/no_catsnout/{0}_no_catsnout.png".format(
timestamp_text)
cv2.putText(self.image, timestamp_text, (2, 14),
cv2.FONT_HERSHEY_SIMPLEX, 0.55, (0, 0, 255), 2)
cv2.putText(self.image, image_text, (2, 32), cv2.FONT_HERSHEY_SIMPLEX,
0.66, (0, 0, 255), 2)
cv2.putText(self.image, "{0:.2f}/3.00".format(trust), (2, 50),
cv2.FONT_HERSHEY_SIMPLEX, 0.52, (0, 0, 255), 2)
cv2.imwrite(filename_mod, self.image)
self.telegram_publisher.publish(filename_mod)
self.lock.release()
return (catsnout_detected, prey_detected)
class Sender:
'''
:param str ip: the IP of the groundstation
:param int port: the receiving port of the groundstation
:param int stream_quality: the JPEG compression rate (0-100)
:param tuple stream_quality: the size of streaming video (width, height)
'''
def __init__(self, ip="192.168.192.101", port=5800, stream_quality=15, stream_size=(352, 256), fps=4.0):
self.ip = ip # IP of ground station
self.port = port # port of socket
self.stream_quality = stream_quality # JPEG quality 0-100
self.stream_size = stream_size
self.streaming = False # flag of streaming video
self.recording = False # flag of recording video
# If TRUE, user requested streamStop(), but waitting __streamThread() stop the thread
self.stoppingStreamThread = False
self.fps = fps
self.sock = socket.socket()
# self.sock.setsockopt(socket.SOL_SOCKET, IN.SO_BINDTODEVICE, "zt2lrwgvd2"+'\0')
print("Connecting to socket %s:%d" % (self.ip, self.port))
self.sock.connect((self.ip, self.port))
# Enable instant reconnection and disable timeout system
self.sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
print("Connected!")
# initialize the camera and grab a reference to the raw camera capture
self.camera = PiCamera()
self.camera.resolution = "1024x768"
self.camera.rotation = 180
self.camera.exposure_compensation = 5 # -25 to 25
self.rawCapture = PiRGBArray(self.camera, size=self.stream_size)
print("Stream quality:", self.stream_quality)
print("Stream size:", self.stream_size)
print("FPS:", self.fps)
def changeQuality(self, qty):
if qty > 100:
qty = 100
if qty < 0:
qty = 0
self.stream_quality = qty
print("Video quality: %i" % qty)
def changeResolution(self, res):
if res == "HD":
self.stream_size = (704, 512)
self.rawCapture = PiRGBArray(self.camera, size=self.stream_size)
self.streamRestart()
print("Change to High Res.")
if res == "SD":
self.stream_size = (352, 256)
self.rawCapture = PiRGBArray(self.camera, size=self.stream_size)
self.streamRestart()
print("Change to Low Res.")
def _videoFileName(self):
now = datetime.datetime.now().strftime('%Y-%m-%d-%H-%M')
return '/data/v-{}.h264'.format(now)
def recordingStart(self):
# Start highres. video recording
if self.recording == False:
print("Recording Start")
self.recording = True
video_file = self._videoFileName()
self.camera.start_recording(video_file)
print("Recording video to "+video_file)
def recordingStop(self):
if self.recording == True:
print("Recording Stop")
self.recording = False
self.camera.stop_recording()
# Continue the recording in the specified output; close existing output
def splitRecording(self):
if self.recording == True:
video_file = self._videoFileName()
self.camera.split_recording(video_file)
print("Split video to "+video_file)
else:
print("No video is recording")
def streamStart(self):
if self.streaming == False:
self.streaming = True
if self.stoppingStreamThread == False:
print("Video stream started")
t = threading.Thread(target=self.__streamThread)
t.start()
else:
print("Video stream start again.")
self.stoppingStreamThread = False
def streamStop(self):
if self.streaming == True and self.stoppingStreamThread == False:
self.streaming = False
self.stoppingStreamThread = True
def streamRestart(self):
self.streamStop()
while self.stoppingStreamThread == True:
pass
self.streamStart()
def increaseExposure(self):
if self.camera.exposure_compensation < 25:
self.camera.exposure_compensation = self.camera.exposure_compensation + 5
print("Exposure compensation:", self.camera.exposure_compensation)
def reduceExposure(self):
if self.camera.exposure_compensation > -25:
self.camera.exposure_compensation = self.camera.exposure_compensation - 5
print("Exposure compensation:", self.camera.exposure_compensation)
def changeFPS(self, data):
if data > 0 and data < 60:
self.fps = data
print("FPS:", self.fps)
# encode the capture image + timestamp
# send it by socket
def __streamThread(self):
for frame in self.camera.capture_continuous(self.rawCapture, format="bgr", use_video_port=True, resize=self.stream_size):
t = time.time()
image = frame.array
encode_param = [IMWRITE_JPEG_QUALITY, self.stream_quality]
result, imgencode = imencode('.jpg', image, encode_param)
data = numpy.array(imgencode)
# Add timestamp
stringData = data.tostring() + str(time.time()).ljust(13)
# Send the size of the data for efficient unpacking
self.sock.sendall(str(len(stringData)).ljust(16).encode())
# Send the data
self.sock.sendall(stringData)
# Clear the stream in preparation for the next frame
self.rawCapture.truncate(0)
# Check if request to stop the stream thread
if self.streaming == False:
self.stoppingStreamThread = False
print("Video stream stop")
break
# Limit FPS to save bandwidth
s = 1/self.fps-(time.time()-t)
if s < 0:
s = 0
time.sleep(s)
def signal():
camera = PiCamera()
camera.resolution = (320, 240)
camera.framerate = 40
rawCapture = PiRGBArray(camera, size=(320, 240))
camera.shutter_speed = 50
camera.iso = 1600
camera.rotation = 180
time.sleep(1)
ret = True
counter = 0
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
# Capture frame-by-frame
img = frame.array[:120][:][::]
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cont = img.copy()
canny = cv2.Canny(gray, 30, 250)
contours, hierarchy = cv2.findContours(canny, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
new_contour = []
counter += 1
hm = 50
get_number = 0
for i in range(0, len(contours)):
epsilon = 0.1 * cv2.arcLength(contours[i], True)
approx = cv2.approxPolyDP(contours[i], epsilon, closed=True)
if (abs(cv2.contourArea(contours[i])) > 0):
# if(abs(cv2.contourArea(contours[i])) > 0 and hierarchy[0][i][0] > -1):
# xm, ym, wm, hm = cv2.boundingRect(contours[hierarchy[0][i][0]])
if ((hm) > 40):
x, y, w, h = cv2.boundingRect(contours[i])
rel = h / w
addh = int(h / 8)
addw = int(w / 8)
if (rel > 1.5 and h > 20 and w > 10 and w < 40 and h < 55 and y > 5 and (y + h) < 235 and x < 220):
# if(rel > 1.3):
# cv2.polylines(cont, [contours[i]], False, (0,255,0), 2)
number = \
cv2.threshold(gray[y - addh:y + h + addh, x - addw:x + w + addw], 30, 255,
cv2.THRESH_BINARY)[1]
try:
if number[0][0].all() == 0:
number = cv2.bitwise_not(number)
nonzeroes = np.count_nonzero(number) / np.prod(number.shape)
if number[0][0].all() != 0 and number[:][0].all() and nonzeroes > 0.65 and nonzeroes < 0.88:
get_number = tesseract(number)
if (show_pictures == "yes"):
cv2.rectangle(cont, (x - addw, y - addh), (x + w + addw, y + h + addh), (0, 0, 255),
1)
cv2.imshow('number', cv2.resize(number, None, fx=5, fy=5))
cv2.putText(cont, "h={} w={} c={}".format(h, w, cv2.isContourConvex(contours[i])),
(x + w + addw + 5, y), cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 255), 1)
cv2.putText(cont,
"r={:.2f} a={} n={}".format(rel, abs(cv2.contourArea(contours[i])), i),
(x + w + addw + 5, y + 15), cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 255), 1)
cv2.putText(cont, "x={} y={}".format(x, y),
(x + w + addw + 5, y + 30), cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 255), 1)
cv2.putText(cont, "number={} l={}".format(get_number, len(approx)),
(x + w + addw + 5, y + 45), cv2.FONT_HERSHEY_PLAIN, 1, (255, 255, 0), 1)
print(str(get_number) + " " + str(
np.count_nonzero(number) / np.prod(number.shape)) + " " + str(
np.count_nonzero(number)) + " " + str(np.prod(number.shape)))
except:
count = 1
if (show_pictures == "yes"):
cv2.imshow('canny', canny)
cv2.imshow('cont', cont)
# cv2.destroyAllWindows()
try:
if int(get_number) > 0:
sound.play(str(int(get_number)))
except:
print("not a number")
rawCapture.truncate(0)
#waitkey is needed, else picture wont display with imshow
if cv2.waitKey(1) & 0xFF == ord('q'):
break
def Follow_Line_Up_Top(camera,
timeout,
testMode=False,
intersectionQueue=[],
intersectionSpeeds=[],
intersectionTimes=[],
intersectionPeriods=[],
robot=loadRobot('ROBOSON.json')):
# Adjusting Robot variables
"""IncreaseTime = False
firstInter = True"""
baseSpeed = robot.speed.base_top
maxSpeed = robot.speed.max_top
minSpeed = robot.speed.min_top
numberOfLinesRequiredForIntersectionMode = robot.number_of_lines_required_for_inersection_mode
# Value used for the binary filter
BIN_CUT = robot.line_finder.binary_cut
# Loading PID values
MultiCoefficient = robot.pid.total_top
PCoefficient = robot.pid.pro_top
DCoefficient = robot.pid.der_top
offlineExponential = robot.pid.off_line_top
#camera = PiCamera()
#camera.color_effects = (128, 128)
cameraResolution = robot.line_finder.resolution
#camera.resolution = (cameraResolution[0], cameraResolution[1])
rawCapture = PiRGBArray(camera,
size=(cameraResolution[0], cameraResolution[1]))
# Check serial port issues
# if there is no serial connection, this function
# simply returned the untouched queue
try:
ser = serial.Serial("/dev/ttyS0", 9600)
ser.flushInput()
serialByteArray = []
except serial.SerialException:
rawCapture.truncate(0)
return intersectionQueue
# Changinig the mode to showing the frames or not
if testMode:
cv2.namedWindow("Original_frame", cv2.WINDOW_NORMAL)
cv2.namedWindow("binary", cv2.WINDOW_NORMAL)
cv2.resizeWindow("binary", 100, 100)
cv2.resizeWindow("Original_frame", 100, 100)
# Reads width of the line from the file
with open('LineWidth.txt') as f:
black_line_width = int(f.readline())
# Minimum width for a fork
fork_min_width = robot.fork_black_line_min_width_multiplier * black_line_width
post_line_width = robot.post_black_line_width * black_line_width
# Kernel setup
n = 3
kernel = np.array([[0, 1 / n, 0] * n])
# Camera initialization and start
camera.capture(rawCapture, format='bgr')
frame = rawCapture.array[:, :, 0]
rawCapture.truncate(0)
# Retrieving the height and the with of the frame
height = len(frame)
width = len(frame[0])
# Sensor lines used for the line follower
# You could change the distances
linesY = np.linspace(
2 / 3 * height - 5, height - 10, 5,
dtype=int) #np.linspace(20, height-5, 5, dtype = int)#
# Required by the camera function to runcate every time
rawCapture.truncate(0)
# List of errors and times for PID plots
if testMode:
errorList = []
timeList = []
# Initializing distances for derivative calculation
currentDeltaX = 0
previousDeltaX = 0
# Initializing times
currentTime = time.time()
frameEndTime = time.time()
frameStartTime = time.time()
# Intersection mode determins the process of the line following
intersectionMode = False
# This value changes based on the last element in the intersection queue
intersectionDirection = None
# Number of lines detectecing a intersection where the left adn right
# spikes are more than a single line
numberOfLinesDetectingIntersection = 0
# Used to defer the intersections if detected too soon
lineFollowingStartTime = time.time()
lastIntersectionDetectionPeriod = intersectionPeriods.pop(0)
lastIntersectionDetectedTime = time.time()
startFollowTopLine = time.time()
# There are two ways to exit this main loop
# 1) The serial connection is lost
# 2) queue of the turns has reached character "X"
# Main for loop starting
# Frame is taken as a 3 channgel grayscaled image
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
if time.time() - startFollowTopLine > timeout:
serialByteArray = [0, 0, 0, 0]
ser.write(serialByteArray)
rawCapture.truncate(0)
return intersectionQueue
# Algorithm start time
algorithmStartTime = time.time()
# Sample frame taken time
frameStartTime = time.time()
frameTakingTime = frameStartTime - frameEndTime
#print("Frame taking time:", frameTakingTime)
# Appending time to list
timeList.append(frameStartTime)
# Assignet previous delta X used for the derivative
previousDeltaX = currentDeltaX
frameArray = frame.array
# Creating a grayscale copy of the frame by taking only one of the channels
# and only including the lines indicated for the line follwoing algorithm
# This step is included to reduce the computer time
frameCopy = frameArray.copy()[linesY]
frameCopyTime = time.time()
#print("transform time", frameCopyTime - algorithmStartTime)
# Taking the kernel of the image
kerneledImage = cv2.filter2D(frameCopy, -1, kernel)[:, :, 0]
kernelTime = time.time()
#print("kernel time: ",kernelTime-frameCopyTime)
# Blurring the image
blurredImage = cv2.GaussianBlur(kerneledImage, (5, 5), 0)
blurTime = time.time()
#print("blur time: ",blurTime-kernelTime)
# Binary filter for the image
ret, binaryImage = cv2.threshold(blurredImage, BIN_CUT, 255,
cv2.THRESH_BINARY)
binTime = time.time()
#print("bin time: ",binTime-blurTime)
# Creating the matrix
pathMatrix = binaryImage
# Median center value
# This value is a cumulative results from all the lines of sensors
currentDeltaX = 0
# list of deltaXs
# A list of the deltaXs indicated by each line
deltaXList = []
# We zero this value to increment it in the for loop to find out
# if the number of sensor lines detecting intersection goes to zero
# in which case we must quit the intersection mode
numberOfLinesDetectingIntersection = 0
if not intersectionMode:
# A loop for the calculations for each line of the sensors
for line_index in range(0, len(linesY)):
lineY = linesY[line_index]
line = pathMatrix[line_index]
# taking the derivative of the lines
dline = diff(line)
if line[0] == 0:
dline[0] = 1
if line[-1] == 0:
dline[-1] = 1
# Here we can either choose the two heighest points
# or we could simple trust that there will only be 2 values
# This will indicate the two edges of the line
#top_two_indices = sorted(range(len(dline)), key = lambda i: dline[i])[-2:]
edgeIndices = [
i for i in range(0, len(dline)) if dline[i] != 0
]
# Case for when two or more edges are detected
if (len(edgeIndices) >= 2):
leftmostEdge = edgeIndices[0]
rightmostEdge = edgeIndices[-1]
lineWidthDetected = rightmostEdge - leftmostEdge
#print("Line width detected: ",lineWidthDetected)
#print(len(edgeIndices))
#print("_______________________________________________")
# Checking to see if a for is detected
if (lineWidthDetected >= fork_min_width
and len(edgeIndices) >= 4):
intersectionDirection = intersectionQueue[0]
numberOfLinesDetectingIntersection += 1
if (intersectionDirection == "L"):
thisLineDeltaX = rightmostEdge - width / 2
elif (intersectionDirection == "R"):
thisLineDeltaX = leftmostEdge - width / 2
elif (intersectionDirection == "X"):
serialByteArray = [100, 0, 100, 0]
ser.write(serialByteArray)
rawCapture.truncate(0)
return intersectionQueue
deltaXList.append(thisLineDeltaX)
elif (lineWidthDetected >= post_line_width):
intersectionDirection = intersectionQueue[0]
numberOfLinesDetectingIntersection += 1
if (intersectionDirection == "L"):
thisLineDeltaX = rightmostEdge - width / 2
elif (intersectionDirection == "R"):
thisLineDeltaX = leftmostEdge - width / 2
elif (intersectionDirection == "X"):
serialByteArray = [100, 0, 100, 0]
ser.write(serialByteArray)
rawCapture.truncate(0)
return intersectionQueue
deltaXList.append(thisLineDeltaX)
else:
# Finding the denter of the line
lineCenterX = int((leftmostEdge + rightmostEdge) / 2)
# Draw circle for showing
# cv2.circle(frame, (lineCenterX, lineY), 2, (255,0,0), -1)
thisLineDeltaX = rightmostEdge - width / 2
deltaXList.append(thisLineDeltaX)
# Case for when only one edge is detected
elif (len(edgeIndices) == 1):
if dline[0] == 0:
onlyEdge = edgeIndices[-1]
if dline[-1] == 0:
onlyEdge = edgeIndices[0]
# Cases for when the edge is to the left or the right
if (onlyEdge >= width / 2):
lineCenterX = onlyEdge + black_line_width / 2
if (onlyEdge < width / 2):
lineCenterX = onlyEdge - black_line_width / 2
thisLineDeltaX = lineCenterX - width / 2
deltaXList.append(thisLineDeltaX)
print("F**K THE POLICE")
# Case for when we are offline
# We will exponentially increase the delta values to return the line
else:
thisLineDeltaX = offlineExponential * (abs(
previousDeltaX)) * (-1 if previousDeltaX < 0 else 1)
deltaXList.append(thisLineDeltaX)
#going into intersection mode
if (numberOfLinesDetectingIntersection >=
numberOfLinesRequiredForIntersectionMode
and time.time() - lastIntersectionDetectedTime >
lastIntersectionDetectionPeriod):
#intersectionMode = True
intersectionDirection = intersectionQueue.pop(0)
intersectionSpeed = intersectionSpeeds.pop(0)
intersectionTime = intersectionTimes.pop(0)
lastIntersectionDetectionPeriod = intersectionPeriods.pop(0)
thisTime = time.time()
goLeft = int(intersectionDirection == 'L')
while (time.time() - thisTime < intersectionTime):
ser.write([
intersectionSpeed, goLeft, intersectionSpeed,
int(not goLeft)
])
ser.write([
int(intersectionSpeed / 3),
int(not goLeft),
int(intersectionSpeed / 3), goLeft
])
time.sleep(0.05)
ser.write([0, 0, 0, 0])
print("intersection mode timestamp", time.time())
print("this intersection period was ",
time.time() - lastIntersectionDetectedTime)
print("__________")
lastIntersectionDetectedTime = time.time()
if intersectionMode:
# Important !
# Pay attention that in this loop now, the
# intersection direction is already determined
# We must have poped it off in the previous iteration from
# the intersectionQueue in order for the state of the machine
# to change to intersection mode
# Also note,
# if the intersection direction indicated "X",
# we should have quitted the program by now
# and there is no need to check
# A loop for the calculations for each line of the sensors
for line_index in range(0, len(linesY)):
lineY = linesY[line_index]
line = pathMatrix[line_index]
# taking the derivative of the lines
dline = diff(line)
if line[0] == 0:
dline[0] = 1
if line[-1] == 0:
dline[-1] = 1
# Here we can either choose the two heighest points
# or we could simple trust that there will only be 2 values
# This will indicate the two edges of the line
#top_two_indices = sorted(range(len(dline)), key = lambda i: dline[i])[-2:]
edgeIndices = [
i for i in range(0, len(dline)) if dline[i] != 0
]
# Case for when two or more edges are detected
if (len(edgeIndices) >= 2):
leftmostEdge = edgeIndices[0]
rightmostEdge = edgeIndices[-1]
lineWidthDetected = rightmostEdge - leftmostEdge
#print("Line width detected: ",lineWidthDetected)
# Checking to see if a for is detected
if (lineWidthDetected >= fork_min_width):
numberOfLinesDetectingIntersection += 1
if (intersectionDirection == "L"):
thisLineDeltaX = rightmostEdge - width / 2
elif (intersectionDirection == "R"):
thisLineDeltaX = leftmostEdge - width / 2
elif (intersectionDirection == "X"):
serialByteArray = [100, 0, 100, 0]
ser.write(serialByteArray)
rawCapture.truncate(0)
return intersectionQueue
deltaXList.append(thisLineDeltaX)
else:
# Finding the denter of the line
lineCenterX = int((leftmostEdge + rightmostEdge) / 2)
# Draw circle for showing
# cv2.circle(frame, (lineCenterX, lineY), 2, (255,0,0), -1)
thisLineDeltaX = lineCenterX - width / 2
deltaXList.append(thisLineDeltaX)
# Case for when only one edge is detected
elif (len(edgeIndices) == 1):
if dline[0] == 0:
onlyEdge = edgeIndices[-1]
if dline[-1] == 0:
onlyEdge = edgeIndices[0]
# Cases for when the edge is to the left or the right
if (onlyEdge >= width / 2):
lineCenterX = onlyEdge + black_line_width / 2
if (onlyEdge < width / 2):
lineCenterX = onlyEdge - black_line_width / 2
thisLineDeltaX = lineCenterX - width / 2
deltaXList.append(thisLineDeltaX)
# Case for when we are offline
# We will exponentially increase the delta values to return the line
else:
thisLineDeltaX = offlineExponential * (abs(
previousDeltaX)) * (-1 if previousDeltaX < 0 else 1)
deltaXList.append(thisLineDeltaX)
if (numberOfLinesDetectingIntersection == 0
): #and time.time() - intersectionStartTime >= 0.3):
print("exiting intersection mode!")
intersectionMode = False
# Takingn the median of the delta Xs
currentDeltaX = statistics.median(deltaXList)
#print("Delta X measured: ", currentDeltaX)
forLoopTime = time.time()
#print("for loop time: ", forLoopTime - binTime)
#
# PID calculations
#
# Finding the derivative time
# The time it took from taking the last frame
derivativeDeltaTime = frameTakingTime + (forLoopTime -
algorithmStartTime)
# finding the derivative
dXdT = (currentDeltaX - previousDeltaX) / derivativeDeltaTime
# Finding the error value given the constants
error_value = int(MultiCoefficient *
(DCoefficient * dXdT + PCoefficient * currentDeltaX))
duty_cycle_left = baseSpeed + error_value
duty_cycle_right = baseSpeed - error_value
#print(duty_cycle_left, duty_cycle_right)
# Reassigning the duty cycle values based on the cutoffs
# Left Wheel
if duty_cycle_left > maxSpeed:
duty_cycle_left = maxSpeed
#print("F**K LEFT")
elif duty_cycle_left < minSpeed:
duty_cycle_left = minSpeed
#print("F**K LEFT")
# Right wheel
if duty_cycle_right > maxSpeed:
duty_cycle_right = maxSpeed
#print("F**K RIGTH")
elif duty_cycle_right < minSpeed:
duty_cycle_right = minSpeed
#print("F**K RIGTH")
calcTime = time.time()
#print("Calculation time: ", calcTime - forLoopTime)
# Serial communication to the BluePill
serialByteArray.append(abs(duty_cycle_left))
if (duty_cycle_left > 0):
serialByteArray.append(1)
else:
serialByteArray.append(0)
serialByteArray.append(abs(duty_cycle_right))
if (duty_cycle_right > 0):
serialByteArray.append(1)
else:
serialByteArray.append(0)
#print("Byte array sent to the BluePill: ",serialByteArray)
# This try cathc block will return the unfinished
# queue in case the serial communication is lost in the middle
try:
ser.write(serialByteArray)
serialByteArray = []
except serial.SerialException:
rawCapture.truncate(0)
return intersectionQueue
serialTime = time.time()
#print("serial time: ", serialTime - calcTime)
# Loop is now complete
key = cv2.waitKey(1)
# if the `q` key was pressed, break from the loop
if key == "q":
break
algorithmEndTime = time.time()
#print("Full algorithm time: ", algorithmEndTime - algorithmStartTime)
# Showing the frame in test mode only
if testMode:
cv2.imshow("Original_frame", frameArray)
cv2.imshow("binary", binaryImage)
# Truncating reqiured for the frames
rawCapture.truncate(0)
# End of this frame
frameEndTime = time.time()
camera.close()
def run(self):
""" initialize the camera and grab a
reference to the raw camera capture """
camera = PiCamera()
camera.resolution = (640, 480)
# camera.hflip = True
# camera.vflip = True
camera.awb_mode = 'off'
# Start off with low gains
rg, bg = (1, 1.8)
camera.awb_gains = (rg, bg)
rawCapture = PiRGBArray(camera)
# allow the camera to warmup
time.sleep(0.1)
# grab an image from the camera
camera.capture(rawCapture, format="bgr")
# Show the full unbalanced image
# image = rawCapture.array
# cv2.imshow("Initial", image)
rawCapture.seek(0)
rawCapture.truncate()
for i in range(10):
camera.capture(rawCapture, format="bgr")
# get the centre section, average RGB channels
imagecentre = rawCapture.array[190:290, 270:370]
# if i == 9:
# cv2.imshow("Image" + str(i), imagecentre)
b, g, r = (np.mean(imagecentre[..., i]) for i in range(3))
print('R:%5.2f, B:%5.2f = (%5.2f, %5.2f, %5.2f)' %
(rg, bg, r, g, b))
if abs(r - g) > 20:
if r > g:
rg -= 0.1
else:
rg += 0.1
if abs(b - g) > 20:
if b > g:
bg -= 0.1
else:
bg += 0.1
if abs(r - g) > 2:
if r > g:
rg -= 0.1
else:
rg += 0.1
if abs(b - g) > 1:
if b > g:
bg -= 0.1
else:
bg += 0.1
camera.awb_gains = (rg, bg)
rawCapture.seek(0)
rawCapture.truncate()
# display the image on screen and wait for a keypress
# camera.capture(rawCapture, format="bgr")
# final_image = rawCapture.array
# cv2.imshow("Final", final_image)
# cv2.waitKey(0)
f = open('rbgains.txt', 'w')
f.write("R:%5.2f\nB:%5.2f\n" % (rg, bg))
f.close()
def aggiungi_persona(self):
d = {}
face_cascade = cv2.CascadeClassifier(
'dataBase/haarcascade_frontalface_default_face.xml')
#carico dizionario
self.carica_personale(d)
#verifica presenza nel database
flag = False
while (flag == False):
print(" ")
nome = input("inserisci nome e cognome: ")
persone = d.values()
if (nome not in persone):
flag = True
else:
print("persona già in database... riprova")
#assegnazione id persona
listaKey = d.keys()
nKey = len(listaKey)
num = 1
while num <= nKey:
if (num not in listaKey):
d[num] = nome #aggiungo persona al database
break
num = num + 1
#izializziamo la picamera
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 32
rawCapture = PiRGBArray(camera, size=(640, 480))
time.sleep(0.1)
#procedura riconoscimento volto
print("")
print("avvicinare il volto alla camera...")
time.sleep(1)
print("3...")
time.sleep(1)
print("2...")
time.sleep(1)
print("1...")
time.sleep(1)
print(" ")
print('Rilevamento di 20 volti...')
print(" ")
#avviamo la picamera e creo database di volti
cont = 0
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
#creazione immagine in array numpy
image = frame.array
#conversione immagine in scala di grigi
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
#troviamo il volto, faces memorizza le coordinate del volto usando il file xml
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
for (x, y, w, h) in faces:
#salviamo il volto ritagliato in scala di grigi
cv2.imwrite(
"dataBase/dataBaseVolti/User." + str(num) + "." +
str(cont + 1) + ".jpg", gray[y:y + h, x:x + w])
cont = cont + 1
print('volto numero ' + str(cont))
rawCapture.truncate(0)
cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 0), 2)
cv2.waitKey(100)
#mostriamo i frame
cv2.imshow("Frame", image)
rawCapture.truncate(0)
cv2.waitKey(1)
if (cont >= 20):
print('Fine rilevamento!')
break
#creo database binario
self.crea_databaseBinario()
cv2.destroyAllWindows()
cv2.waitKey(1)
cv2.waitKey(1)
cv2.waitKey(1)
cv2.waitKey(1)
camera.close()
self.salva_personale(d)
d = {}
domanda = input("Vuoi aggiungere un'altra persona? Y/N :")
if (domanda == 'Y' or domanda == 'y'):
self.aggiungi_persona()
else:
print("")
print("fine opzioni")
ang = 90 + ang
setpoint = 100 ####### setpoint is used to make origin point out of x_min
error = int(x_min - setpoint)
ang = int(ang)
Motor_speed(45, (error * kp) + (ang * ap))
box = cv2.boxPoints(
blackbox
) #obtaining points that serounds the minAreaRect to be able to draw it as a contour in the image
box = np.int0(box)
cv2.drawContours(
image, [box], 0, (0, 0, 255),
3) #draw the contour to the image so that we can see it
cv2.putText(image, str(ang), (10, 40), cv2.FONT_HERSHEY_SIMPLEX, 1,
(0, 0, 255), 2)
cv2.putText(image, str(error), (10, 100), cv2.FONT_HERSHEY_SIMPLEX, 1,
(255, 0, 0), 2)
cv2.line(image, (int(x_min), 90), (int(x_min), 110), (255, 0, 0), 3)
cv2.imshow("orginal with line", image_Binary
) #shows the variale image that holds the array of the image.
rawCapture.truncate(
0
) # the function truncate is used here to empty the buffer sothat the next fram can be written to it.
key = cv2.waitKey(1) & 0xFF # read a key
if key == ord(
"q"
): # if the pressed key is the charachter q then the loop breaks and finishes
break
GPIO.output(40, GPIO.LOW) # turn the light of the camera off
time.sleep(0.1)
for frameBGR in cam.capture_continuous(raw, format="bgr", use_video_port=True):
imgBGR = frameBGR.array
imgBW = cv2.cvtColor(imgBGR, cv2.COLOR_BGR2GRAY)
listFace = faceCascade.detectMultiScale(
imgBW,
scaleFactor=1.1,
minNeighbors=5,
minSize=(30, 30),
#v2.X
#flags=cv2.cv.CV_HAAR_SCALE_IMAGE
#v3.1+
flags=cv2.CASCADE_SCALE_IMAGE)
for (x, y, w, h) in listFace:
cv2.rectangle(imgBGR, (x, y), (x + w, y + h), (255, 255, 0), 4)
print len(listFace)
cv2.imshow('Video', imgBGR)
key = cv2.waitKey(1) & 0xFF
raw.truncate(0) # Clear stream
if key == ord("q"):
break
class Camera(PiCamera):
def __init__(self,
sonar=None,
clientAddr=None,
clientPort=5555,
resolution=(640, 480)):
PiCamera.__init__(self)
self.resolution = resolution
self.mode = None
self.sonar = sonar
self.rawCapture = PiRGBArray(self)
self.rawCapture.truncate(0)
self.rawCapture.seek(0)
# Start stream
#self.stream()
def stream(self, addr='192.168.0.169', port=5555):
'''Send live stream'''
# Now connect the camera socket
context = zmq.Context()
footageSocket = context.socket(zmq.PUB)
footageSocket.connect('tcp://{}:{}'.format(addr, port))
self.rawCapture.truncate(0)
self.rawCapture.seek(0)
#font = cv2.FONT_HERSHEY_SIMPLEX
for frame in self.capture_continuous(output=self.rawCapture,
format="bgr",
use_video_port=True):
image = frame.array
# draw cross on image
#cv2.line(image,(300,240),(340,240),(128,255,128),1)
#cv2.line(image,(320,220),(320,260),(128,255,128),1)
if self.mode == 'opencv':
pass #TODO: impliment open cv modes
'''
try:
if self.sonar.distance < 8:
cv2.putText(image,'%s m'%str(round(self.sonar.distance,2)),
(40,40), font, 0.5,(255,255,255),
1,cv2.LINE_AA)
except:
pass
'''
# send image to client
encoded, buffer = cv2.imencode('.jpg', image)
jpg_as_text = base64.b64encode(buffer)
footageSocket.send(jpg_as_text)
#footageSocket.send(image)
#send_array(footageSocket,image)
self.rawCapture.truncate(0)
self.rawCapture.seek(0)
class Tracker:
'''
iFollow robot class
'''
def __init__(self, colors, frameSize):
"""
Args: List of 2 color tuples: [(0,0,0),(255,255,255)]
"""
self.colors = colors
# capturing video through webcam
self.camera = PiCamera()
self.rawCapture = PiRGBArray(self.camera)
self.status = "Running"
self.minRadius = 20
self.cameraWidth = frameSize
def __str__(self):
'''
Object Representation
'''
print("Ok!")
def prepareColors(self):
colors = self.colors
Min = colors[0]
Max = colors[1]
return (np.array(list(Min), np.uint8), np.array(list(Max), np.uint8))
def detectObject(self, mask, frame):
contours = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)[-2]
if len(contours) > 0:
c = max(contours, key=cv2.contourArea)
((x, y), radius) = cv2.minEnclosingCircle(c)
M = cv2.moments(c)
center = (int(M["m10"] / M["m00"]), int(M["m01"] / M["m00"]))
# only proceed if the radius meets a minimum size
if radius > self.minRadius:
# draw the circle and centroid on the frame,
# then update the list of tracked points
cv2.circle(frame, (int(x), int(y)), int(radius),
(255, 255, 255), 2)
cv2.circle(frame, center, 2, (0, 255, 0), -1)
return x
return None
def createMask(self, hsv, frame):
masks = []
color = self.prepareColors()
# finding the range of red,blue and yellow color in the image
mask = cv2.inRange(hsv, color[0], color[1])
mask = cv2.erode(mask, None, iterations=2)
mask = cv2.dilate(mask, None, iterations=2)
# Morphological transformation, Dilation
kernal = np.ones((5, 5), "uint8")
mask = cv2.dilate(mask, kernal)
#mask = cv2.bitwise_and(frame, frame, mask = mask)
return mask
def getCoordinate(self, blur, frame=np.array([False, False])):
# keep looping
# grab the current frame
if not frame.all():
stream = io.BytesIO()
camera = self.camera
camera.resolution = (400, 400)
camera.start_preview()
stream = PiRGBArray(camera)
camera.capture(stream, format='bgr')
# At this point the image is available as stream.array
frame = stream.array
#image = self.rawCapture.array
# resize the frame, blur it, and convert it to the HSV color space
frame = imutils.resize(frame, width=self.cameraWidth)
frame = cv2.flip(frame, 1)
blurred = cv2.GaussianBlur(frame, (11, 11), 0) if blur else frame
hsv = cv2.cvtColor(blurred, cv2.COLOR_BGR2HSV)
mask = self.createMask(hsv, frame)
self.rawCapture.truncate(0)
# Track each color:
return self.detectObject(mask, frame)
def talker():
#rospy.init_node('green_talker_rpi', anonymous=True)
#pub = rospy.Publisher('green_chatter_rpi', ballCor, queue_size=50)
r = rospy.Rate(10) #10hz
# Initialize the Green ball color
#greenLower = (29,86,100)
#greenUpper = (60,255,255)
global msg
pts = deque(maxlen=64)
#Grab webCam
#camera = cv2.VideoCapture(0)
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 62
rawCapture = PiRGBArray(camera, size=(640, 480))
time.sleep(0.1)
# msg.heading="RPI-1 talker started"
# msg.color="RED"
# msg.x = 0
# msg.y = 0
# msg.radius = 0
# msg.upperH = greenUpper[0]
# msg.upperS = greenUpper[1]
# msg.upperV = greenUpper[2]
# msg.lowerH = greenLower[0]
# msg.lowerS = greenLower[0]
# msg.lowerV = greenLower[0]
global detected
detected = False
# keep looping
for myframe in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
# grab the current frame
#(grabbed, frame) = camera.read()
frame = myframe.array
# resize the frame, blur it, and convert it to the HSV
# color space
frame = imutils.resize(frame, width=600)
# blurred = cv2.GaussianBlur(frame, (11, 11), 0)
hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
# construct a mask for the color "green", then perform
# a series of dilations and erosions to remove any small
# blobs left in the mask
mask = cv2.inRange(hsv, greenLower, greenUpper)
mask = cv2.erode(mask, None, iterations=2)
mask = cv2.dilate(mask, None, iterations=2)
# find contours in the mask and initialize the current
# (x, y) center of the ball
cnts = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)[-2]
center = None
# only proceed if at least one contour was found
if len(cnts) > 0:
# find the largest contour in the mask, then use
# it to compute the minimum enclosing circle and
# centroid
c = max(cnts, key=cv2.contourArea)
((x, y), radius) = cv2.minEnclosingCircle(c)
M = cv2.moments(c)
center = (int(M["m10"] / M["m00"]), int(M["m01"] / M["m00"]))
#rospy.loginfo(radius)
# only proceed if the radius meets a minimum size
if radius > 5:
# draw the circle and centroid on the frame,
# then update the list of tracked points
cv2.circle(frame, (int(x), int(y)), int(radius),
(0, 255, 255), 2)
cv2.circle(frame, center, 5, (0, 0, 255), -1)
msg.x=int(x)
msg.y=int(y)
msg.radius=int(radius)
#rospy.loginfo(detected)
if detected == False:
rospy.loginfo(" "+ msg.color+" ball detected ")
detected = True
#detected = True
else:
#rospy.loginfo(detected)
if detected == True:
rospy.loginfo("Can't find the ball")
rospy.loginfo("Sending request to neighbours to find "+msg.color+" ball.")
#rospy.loginfo(msg)
pub.publish(msg)
detected = False
#print "Publishing to Lap cam"
#r.sleep()
#return 0
# update the points queue
#pts.appendleft(center)
# loop over the set of tracked points
#for i in xrange(1, len(pts)):
# if either of the tracked points are None, ignore
# them
#if pts[i - 1] is None or pts[i] is None:
# continue
# otherwise, compute the thickness of the line and
# draw the connecting lines
#thickness = int(np.sqrt(64 / float(i + 1)) * 2.5)
#cv2.line(frame, pts[i - 1], pts[i], (0, 0, 255), thickness)
# show the frame to our screen
#cv2.imshow("Frame", frame)
rawCapture.truncate(0)
list_results = [[0] * 7 for x in xrange(num_pics + 1)]
list_results[0] = [
'Trial',
't (take pic)',
't (save pic)',
't (SIFT)',
'Found? (SIFT)',
't (ArUco)',
'Found? (ArUco)',
]
for i in range(num_pics):
print '%0.2f: Trial %03d' % ((time.time() - time_start), i)
list_results[i + 1][0] = i
time_meas = time.time()
rawCapt.truncate(0)
camera.capture(rawCapt, format='bgr', use_video_port=True)
pic = rawCapt.array
list_results[i + 1][1] = time.time() - time_meas
time_meas = time.time()
path = 'pic_%s_Trial%03d.jpg' % (filename, i)
cv2.imwrite(path, pic)
list_results[i + 1][2] = time.time() - time_meas
list_results[i + 1][3], list_results[i + 1][4] = find_target(
pic, 'sift')
list_results[i + 1][5], list_results[i + 1][6] = find_target(
pic, 'aruco')
def cv_Demo2():
rawCapture = PiRGBArray(camera)
camera.iso = 100
# Wait for the automatic gain control to settle
time.sleep(2)
global angle
global marker_distance_cm
# Now fix the values
camera.shutter_speed = camera.exposure_speed
camera.exposure_mode = 'off'
g = camera.awb_gains
camera.awb_mode = 'off'
camera.awb_gains = g
# allow the camera to warmup
time.sleep(0.1)
camera.framerate = 32
aruco_dictionary = aruco.Dictionary_get(aruco.DICT_6X6_250)
parameters = aruco.DetectorParameters_create()
timer = 0
xwidth = 54 * (math.pi / 180)
ywidth = 41 * (math.pi / 180)
marker_flag = 1
# start continuous picture taking
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
marker_found = False
# convert to grayscale
image = frame.array
cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
corners, ids, rejectedImgPoints = aruco.detectMarkers(
image, aruco_dictionary, parameters=parameters)
if type(ids) == numpy.ndarray:
print("Marker found")
marker_found = True
output_array = numpy.ndarray(len(ids))
index = 0
for i in ids:
for j in i:
output_array[index] = j
print("Marker Indexes:", end=" ")
print(j)
#Determine angle of marcker form the center of the camera
xcenter = int(
(corners[0][0][0][0] + corners[0][0][2][0]) / 2)
ycenter = int(
(corners[0][0][0][1] + corners[0][0][2][1]) / 2)
yheight = abs(
int(corners[0][0][0][1] - corners[0][0][2][1]))
marker_distance_ft = 490 / yheight
print(yheight)
print("Marker Distance in ft: ", end="")
print(marker_distance_ft)
print("Rounding: ", end="")
print(round(marker_distance_ft))
print(marker_distance_cm)
# marker_distance_cm = [int(ele) for ele in str(marker_distance_cm) if ele.isdigit()]
#print("Marker Distacne in m: ", end="")
#print(marker_distance_m)
imageMiddleX = image.shape[1] / 2
xdist = (xcenter - image.shape[1] / 2)
ydist = (ycenter - image.shape[0] / 2)
xangle = (xdist / image.shape[1]) * xwidth
yangle = (ydist / image.shape[0]) * ywidth
# Calculate the angle from the z-axis to the center point
# First calculate distance (in pixels to screen) on z-axis
a1 = xdist / math.tan(xangle)
a2 = ydist / math.tan(yangle)
a = (a1 + a2) / 2
distance = math.sqrt(pow(xdist, 2) + pow(ydist, 2))
#Calculate final angle for each Aruco
angle = math.atan2(distance, a) * (180 / math.pi)
print("Correction: ", end="")
marker_distance_ft = marker_distance_ft * 0.96 - (
-2.5244 * pow(angle * math.pi / 180, 2) + 0.027)
print(marker_distance_ft)
marker_distance_cm = marker_distance_ft * 30.48 * 100
marker_distance_cm = int(marker_distance_cm)
if (xcenter > imageMiddleX):
angle *= -1
if (angle < -3.8 and angle > -4.3):
angle += 4
else:
angle += 4
if (angle > 0):
angle -= 2
display_angle = str(angle)
print("Angle from camera:", angle, "degrees")
lcd.message = ("Beacon Detected \nAngle: %s" %
(display_angle))
marker_flag = 1
index += 1
print()
break
rawCapture.truncate(0)
if marker_found == False:
print("No markers found")
if marker_flag == 1:
lcd.clear()
marker_flag = 0
lcd.message = ("Beacon Not\nDetected")
else:
break
from picamera.array import PiRGBArray
from picamera import PiCamera
import time
import cv2 # 导入需要的库
camera = PiCamera()
camera.resolution = (640, 480) # 设置分辨率
camera.framerate = 32 # 设置帧率
rawCapture = PiRGBArray(camera, size=(640, 480))
time.sleep(0.1) # 等待摄像头模块初始化
for frame in camera.capture_continuous(rawCapture,
format="bgr",
use_video_port=True):
image = frame.array
cv2.imshow("Frame", image) # 显示图像
key = cv2.waitKey(1) & 0xFF # 等待按键
rawCapture.truncate(0) # 准备下一副图像
if key == ord("q"):
break
class Camera(): #Camera object - handles everything image related
def __init__(self):
print("initializing Pi Camera...")
self.cap = PiCamera() #create Raspberry Pi camera module
self.cap.resolution = (640, 480) #define capture resolution
self.cap.rotation = 90 #compensate for camera orientation
self.cap.framerate = 30 #define capture rate
self.cap.led = False #diasable camera status indicator LED
self.rawCapture = PiRGBArray(self.cap,
size=(640,
480)) #image numpy array object
self.frame = None
self.captureThread = Thread(
target=self.__captureFrame
) #frame retrieval thread - reduce I/O latency
self.captureThread.daemon = True
self.captureThread.do_run = True
self.captureThread.start()
self.net = cv2.dnn.readNetFromCaffe(
"camera/deploy.prototext.txt",
"camera/res10_300x300_ssd_iter_140000.caffemodel")
self.minimumConfidence = 0.5
print("done")
def __captureFrame(self): #frame retrieval thread
for capture in self.cap.capture_continuous(
self.rawCapture, format='bgr',
use_video_port=True): #capture frames in infinte iterator
self.frame = capture.array
self.rawCapture.truncate(0)
if (self.captureThread.do_run == False):
break
def averageGraySpace(
self, grayFrame
): #calculate average light level from frame in gray space
average = np.mean(grayFrame.flatten())
return average
def getFrame(self): #retrieve frame from camera as an array
return self.frame
def convertGray(self, frame): #convert RGB frame to gray space
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
return gray
def getFaces(
self,
frame): #find faces in frame array with haarcascade classifier
self.net.setInput(
cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)), 1.0,
(300, 300), (104.0, 177.0, 123.0)))
faces = self.net.forward()
return faces
def saveFrame(self, frame, name="image"): #save frame array to disk
cv2.imwrite(name, frame)
def __del__(self): #close safely
print("stopping Pi Camera...")
self.captureThread.do_run = False
self.captureThread.join()
self.cap.close()
print("done")
class VideoStream:
"""Camera object"""
def __init__(self, resolution=(640, 480), framerate=30, PiOrUSB=2, src=0):
# Create a variable to indicate if it's a USB camera or PiCamera.
# PiOrUSB = 1 will use PiCamera. PiOrUSB = 2 will use USB camera.
self.PiOrUSB = PiOrUSB
if self.PiOrUSB == 1: # PiCamera
# Import packages from picamera library
from picamera.array import PiRGBArray
from picamera import PiCamera
# Initialize the PiCamera and the camera image stream
self.camera = PiCamera()
self.camera.resolution = resolution
self.camera.framerate = framerate
self.rawCapture = PiRGBArray(self.camera, size=resolution)
self.stream = self.camera.capture_continuous(self.rawCapture,
format="bgr",
use_video_port=True)
# Initialize variable to store the camera frame
self.frame = []
if self.PiOrUSB == 2: # USB camera
# Initialize the USB camera and the camera image stream
self.stream = cv2.VideoCapture(src)
ret = self.stream.set(3, resolution[0])
ret = self.stream.set(4, resolution[1])
#ret = self.stream.set(5,framerate) #Doesn't seem to do anything so it's commented out
# Read first frame from the stream
(self.grabbed, self.frame) = self.stream.read()
# Create a variable to control when the camera is stopped
self.stopped = False
def start(self):
# Start the thread to read frames from the video stream
Thread(target=self.update, args=()).start()
return self
def update(self):
if self.PiOrUSB == 1: # PiCamera
# Keep looping indefinitely until the thread is stopped
for f in self.stream:
# Grab the frame from the stream and clear the stream
# in preparation for the next frame
self.frame = f.array
self.rawCapture.truncate(0)
if self.stopped:
# Close camera resources
self.stream.close()
self.rawCapture.close()
self.camera.close()
if self.PiOrUSB == 2: # USB camera
# Keep looping indefinitely until the thread is stopped
while True:
# If the camera is stopped, stop the thread
if self.stopped:
# Close camera resources
self.stream.release()
return
# Otherwise, grab the next frame from the stream
(self.grabbed, self.frame) = self.stream.read()
def read(self):
# Return the most recent frame
return self.frame
def stop(self):
# Indicate that the camera and thread should be stopped
self.stopped = True
def main():
parser = argparse.ArgumentParser()
parser.add_argument("--model",
help="File path of Tflite model.",
required=True)
parser.add_argument("--label",
help="File path of label file.",
required=True)
parser.add_argument("--top_k", help="keep top k candidates.", default=3)
parser.add_argument("--threshold",
help="threshold to filter results.",
default=0.5,
type=float)
parser.add_argument("--width",
help="Resolution width.",
default=640,
type=int)
parser.add_argument("--height",
help="Resolution height.",
default=480,
type=int)
args = parser.parse_args()
# Initialize window.
cv2.namedWindow(
WINDOW_NAME,
cv2.WINDOW_GUI_NORMAL | cv2.WINDOW_AUTOSIZE | cv2.WINDOW_KEEPRATIO)
cv2.moveWindow(WINDOW_NAME, 100, 200)
# Initialize engine.
engine = DetectionEngine(args.model)
labels = read_label_file(args.label) if args.label else None
# Generate random colors.
last_key = sorted(labels.keys())[len(labels.keys()) - 1]
colors = visual.random_colors(last_key)
elapsed_list = []
resolution_width = args.width
rezolution_height = args.height
with picamera.PiCamera() as camera:
camera.resolution = (resolution_width, rezolution_height)
camera.framerate = 30
_, width, height, channels = engine.get_input_tensor_shape()
rawCapture = PiRGBArray(camera)
# allow the camera to warmup
time.sleep(0.1)
try:
for frame in camera.capture_continuous(rawCapture,
format="rgb",
use_video_port=True):
rawCapture.truncate(0)
# input_buf = np.frombuffer(stream.getvalue(), dtype=np.uint8)
image = frame.array
im = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
input_buf = PIL.Image.fromarray(image)
# Run inference.
ans = engine.detect_with_image(
input_buf,
threshold=args.threshold,
keep_aspect_ratio=False,
relative_coord=False,
top_k=args.top_k,
)
elapsed_ms = engine.get_inference_time()
# Display result.
if ans:
for obj in ans:
label_name = "Unknown"
if labels:
label_name = labels[obj.label_id]
caption = "{0}({1:.2f})".format(label_name, obj.score)
# Draw a rectangle and caption.
box = obj.bounding_box.flatten().tolist()
visual.draw_rectangle(im, box, colors[obj.label_id])
visual.draw_caption(im, box, caption)
# Calc fps.
elapsed_list.append(elapsed_ms)
avg_text = ""
if len(elapsed_list) > 100:
elapsed_list.pop(0)
avg_elapsed_ms = np.mean(elapsed_list)
avg_text = " AGV: {0:.2f}ms".format(avg_elapsed_ms)
# Display fps
fps_text = "{0:.2f}ms".format(elapsed_ms)
visual.draw_caption(im, (10, 30), fps_text + avg_text)
# display
cv2.imshow(WINDOW_NAME, im)
if cv2.waitKey(10) & 0xFF == ord("q"):
break
finally:
camera.stop_preview()
# When everything done, release the window
cv2.destroyAllWindows()
def showVideo(self,click_allow_arg):
lower_IR = np.array([0,0,50]) # lower bound of CV filter
upper_IR = np.array([255,255,255]) # upper bound of CV filter
rawCapture = PiRGBArray(self.cam_handle, size=self.cam_res) # capture to array
m = PyMouse() # make mouse object
prev_maxLoc = (0,0) # init previous maxLoc
try:
for frame in self.cam_handle.capture_continuous(rawCapture, format = 'bgr', use_video_port=True):
image = frame.array # get the array values for that frame
mask = cv2.inRange(image, lower_IR, upper_IR) # mask according to lower/upper filter values
(minVal, maxVal, minLoc, maxLoc) = cv2.minMaxLoc(mask) # find min/max values of image
true_pnt = self.get_true_point(maxLoc)
print true_pnt #DEBUG INFO PRINTOUT
adjusted_maxx = int(((true_pnt[0]) * self.output_res[0])/(self.screen_res[0])) # take the distance from the edge and perform dimensional analysis
adjusted_maxy = int(((true_pnt[1]) * self.output_res[1])/(self.screen_res[1])) # take the distance from the edge and perform dimensional analysis
if click_allow_arg: # if user wants to have clicks when IR found
if maxLoc != (0,0): # if not on the default location
m.press(adjusted_maxx,adjusted_maxy, 1) # press the "mouse" button at location found
elif prev_maxLoc != (0,0) and maxLoc == (0,0): # if the current value is the default and the last value wasn't the default, release
m.release(prev_maxLoc[0],prev_maxLoc[1], 1) # release the "mouse" button
else: # only move the mouse, no clicking
if maxLoc != (0,0): # if not on the default location
m.move(adjusted_maxx,adjusted_maxy) # move the mouse
prev_maxLoc = (adjusted_maxx,adjusted_maxy) # set previous to be current max loc
cv2.circle(image, maxLoc, 21, (255,0,0), 2) # place circle on brightest spot
cv2.imshow("TestWindow", cv2.resize(image,(160,120))) # show the image in a tiny window
cv2.waitKey(1) & 0xFF # needed for preview
rawCapture.seek(0) # return to the 0th byte
rawCapture.truncate() # clear array for next frame
print "Mouse at: ", m.position() #DEBUG INFO PRINTOUT
except KeyboardInterrupt: # used to quit the function
rawCapture.seek(0) # return to the 0th byte
rawCapture.truncate() # clear array for next frame
def main():
print("Starting Program")
# Init camera
camera = PiCamera()
camera.resolution = (640,480)
camera.framerate = 24
rawCapture = PiRGBArray(camera, size=(640,480))
# allow camera to warmup
time.sleep(0.1)
# capture frames from camera
for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
image = frame.array
cv2.imshow("Frame", image)
key = cv2.waitKey(1) & 0xFF
rawCapture.truncate(0)
if key == ord("q"):
break
print("Completed!")
return 0
