def start():
'''
'''
#Load splash screen
splScr = splash()
found = []
#find connected cameras
for num in range(10):
cam = VideoCapture(num)
cam.open
#show progress bar 'movement' while the main program find cameras
splScr.update()
if not cam.read()[0]:
del(cam)
else:
cam.release()
found.append(num)
while gtk.events_pending():
gtk.main_iteration()
#destroy splash screen when all cameras are finded
splScr.destroy()
print 'connected cameras:', len(found)
#run main program
main_gui(found)
gtk.main()
return
class Camera(object):
def get_settings(self):
if not hasattr(self, '_video_capture'):
raise Exception("Start video capture before getting settings")
settings = []
for prop in global_camera_properties:
prop_value = self._video_capture.get(prop['value'])
if prop_value >= 0:
settings.append({'name': prop['name'], 'value': prop_value})
return settings
def set_setting(self, setting, value):
if not hasattr(self, '_video_capture'):
raise Exception("Start video capture before setting a setting")
setting_id = filter(lambda x: x['name'] == setting, global_camera_properties)
if len(setting_id) == 1:
setting_id = setting_id[0]['value']
else:
raise Exception("Setting {} not available".format(setting))
self._video_capture.set(setting_id, value)
def read(self):
(retVal, image) = self._video_capture.read()
return image
def start(self):
self._video_capture = VideoCapture(0)
self.shape = self.read().shape
def stop(self):
self._video_capture.release()
class Camera_Capture():
"""
VideoCapture without uvc control using cv2.VideoCapture
"""
def __init__(self,src_id,size=(640,480),fps=None,timebase=None):
self.controls = None
self.cvId = src_id
self.name = "VideoCapture"
self.controls = None
###add cv videocapture capabilities
self.capture = VideoCapture(src_id)
self.set_size(size)
if timebase == None:
logger.debug("Capture will run with default system timebase")
self.timebase = c_double(0)
elif isinstance(timebase,c_double):
logger.debug("Capture will run with app wide adjustable timebase")
self.timebase = timebase
else:
logger.error("Invalid timebase variable type. Will use default system timebase")
self.timebase = c_double(0)
def get_frame(self):
s, img = self.capture.read()
timestamp = time()
return Frame(timestamp,img)
def set_size(self,size):
width,height = size
self.capture.set(3, width)
self.capture.set(4, height)
def get_size(self):
return self.capture.get(3), self.capture.get(4)
def set_fps(self,fps):
self.capture.set(5,fps)
def get_fps(self):
return self.capture.get(5)
def get_now(self):
return time()
def create_atb_bar(self,pos):
size = 0,0
return size
def kill_atb_bar(self):
pass
def close(self):
pass
class FileCapture():
"""
simple file capture that can auto_rewind
"""
def __init__(self,src):
self.auto_rewind = True
self.controls = None #No UVC controls available with file capture
# we initialize the actual capture based on cv2.VideoCapture
self.cap = VideoCapture(src)
timestamps_loc = os.path.join(src.rsplit(os.path.sep,1)[0],'eye_timestamps.npy')
logger.info("trying to load timestamps with video at: %s"%timestamps_loc)
try:
self.timestamps = np.load(timestamps_loc).tolist()
logger.info("loaded %s timestamps"%len(self.timestamps))
except:
logger.info("did not find timestamps")
self.timestamps = None
self._get_frame_ = self.cap.read
def get_size(self):
return self.cap.get(3),self.cap.get(4)
def set_fps(self):
pass
def get_fps(self):
return None
def read(self):
s, img =self._get_frame_()
if self.auto_rewind and not s:
self.rewind()
s, img = self._get_frame_()
return s,img
def get_frame(self):
s, img = self.read()
if self.timestamps:
timestamp = self.timestamps.pop(0)
else:
timestamp = time()
return Frame(timestamp,img)
def rewind(self):
self.cap.set(1,0) #seek to the beginning
def create_atb_bar(self,pos):
return 0,0
def kill_atb_bar(self):
pass
def close(self):
pass
class CaptureSource:
def __init__(self):
self._video = None
self._image = None
def get_resolution(self):
if (self._video):
return (int(self._video.get(CAP_PROP_FRAME_WIDTH)),
int(self._video.get(CAP_PROP_FRAME_HEIGHT)))
def camera(self, num_cams):
cam = Cameras()
cam.check_cameras(num_cams)
self._video = cam.show_and_select_camera()
def video(self, filename):
self._video = VideoCapture(filename)
def image(self, filename):
self._image = filename
def get_frame(self):
if self._video:
retval, frame = self._video.read()
return retval, frame
return True, imread(self._image)
def __init__(self, camera, dic=None):
self.camera = int(camera)
self.cap = VideoCapture(self.camera)
if dic:
for propID, value in dic.iteritems():
self.cap.set(propID, value)
first_frame = self.frame()
def __init__(self, cam_num = -1, resolution = (640, 480)):
'''
create camera object
cam_num device number (integer)
resolution image resolution (tuple width x height)
'''
self.device = cam_num
self.resolution = resolution
self.BGRimage = []
self.HSVimage = []
self.RGBimage = []
self.FPS = [0, 0]
self.__avr = 0
#assign and open device
self.__capture = VideoCapture(cam_num)
self.__capture.set(CV_CAP_PROP_FRAME_WIDTH,resolution[0])
self.__capture.set(CV_CAP_PROP_FRAME_HEIGHT,resolution[1])
self.__capture.open
self.__flag = False
t0 = time()
self.__flag, self.BGRimage = self.__capture.read()
self.FPS[0] = 1/(time()-t0)
self.FPS[1] = self.FPS[0]
self.__avr = self.FPS[0]
print "camera", self.device, "ready @", self.FPS[0], "fps"
return
def caputure():
# open Camera
cam = VideoCapture(0)
if not cam.isOpened():
LOGGER.debug('FAILED to open camera!!!')
return None
# capture image
for i in range(100):
status, img = cam.read()
if not status:
LOGGER.debug('FAiLED to capture image!!!')
return None
cam.release()
return img
def __init__(self, mouse, n=1, data_dir='.', diff_thresh=80, resample=1, translation_max=50, smoothing_kernel=19, consecutive_skip_threshold=0.08, selection_from=[], point_mode='auto'):
self.mouse = mouse
self.n = n
self.data_dir = data_dir
# Parameters (you may vary)
self.diff_thresh = diff_thresh
self.resample = resample
self.translation_max = translation_max
self.kernel = smoothing_kernel
# Parameters (you should not vary)
self.cth1 = 0
self.cth2 = 0
plat = sys.platform
if 'darwin' in plat:
self.fourcc = CV_FOURCC('m','p','4','v')
elif plat[:3] == 'win':
self.fourcc = 1
else:
self.fourcc = -1
self.fh = FileHandler(self.data_dir, self.mouse, self.n)
self.framei = 0
self.load_time()
self.consecutive_skip_threshold = (self.fs/self.resample) * consecutive_skip_threshold
self.load_background()
self.height, self.width = self.background.shape
self.mov = VideoCapture(self.fh.get_path(self.fh.TRIAL, self.fh.MOV))
self.mov.read();self.time=self.time[1:]
#self.get_frame(self.mov,n=40) #MUST ADJUST TIME IF USING THIS
self.load_pts(mode=point_mode)
self.make_rooms()
def __init__(self, src,size=(640,480)):
self.controls = None
self.cvId = src
###add cv videocapture capabilities
self.cap = VideoCapture(src)
self.set_size(size)
self.read = self.cap.read
def __init__(self, cam,size=(640,480)):
super(CameraCapture, self).__init__(cam)
###add cv videocapture capabilities
self.cap = VideoCapture(self.cvId)
self.set_size(size)
self.read = self.cap.read
def __init__(self, title="Video Stream"):
''' Uses OpenCV 2.3.1 method of accessing camera '''
self.title = title
self.cap = VideoCapture(0)
self.prev = self.get_frame()
self.frame = self.get_frame()
namedWindow(title, 1)
class Camera(object):
"""
The class responsible for communicating with the actual camera and
getting images from it.
Attributes:
cam: An instance of an openCV VideoCapture.
"""
def __init__(self, device_num):
"""
Uses a device num in case the system has multiple cameras attached.
"""
self.cam = VideoCapture(device_num)
def get_image(self):
"""
Grab a frame from the camera. The cameraCommunicator is the caller,
and is responsible for lighting and location. The filename of the
image is returned.
Raises:
FatalCameraException: An image was not taken successfully.
"""
#create the systematic filename
timestamp = datetime.datetime.now()
filename = utils.get_image_dir() + str(timestamp.date()) + \
str(timestamp.hour) + str(timestamp.minute) + \
str(timestamp.second) + '.jpg'
#A series of reads to allow the camera to adjust to lighting
self.cam.read()
self.cam.read()
self.cam.read()
self.cam.read()
#only the last is saved
success, image = self.cam.read()
if not success:
raise FatalCameraException()
else:
imwrite(filename, image)
return timestamp, filename
def __init__(self, mouse, mode, data_directory='.', diff_thresh=100, resample=8, translation_max=100, smoothing_kernel=19, consecutive_skip_threshold=2, selection_from=[]):
self.mouse = mouse
self.data_dir = data_directory
self.mode = mode
self.selection_from = selection_from
# Parameters (you may vary)
self.diff_thresh = diff_thresh
self.resample = resample
self.translation_max = translation_max
self.kernel = smoothing_kernel
self.consecutive_skip_threshold = (37./self.resample) * consecutive_skip_threshold
# Parameters (you should not vary)
self.duration = 1
self.cth1 = 0
self.cth2 = 0
plat = sys.platform
if 'darwin' in plat:
self.fourcc = CV_FOURCC('m','p','4','v')
elif plat[:3] == 'win':
self.fourcc = 1
else:
self.fourcc = -1
fh = FileHandler(self.data_dir, self.mouse)
self.background_name = fh[mode][BACKGROUND][NAME]
self.background_dir = fh[mode][BACKGROUND][DIR]
self.trial_name = fh[mode][TRIAL][NAME]
self.trial_dir = fh[mode][TRIAL][DIR]
self.background, self.background_image = self.load_background()
self.height, self.width = np.shape(self.background)
timefile = os.path.join(self.trial_dir, self.trial_name+'-timestamps.json')
self.time = json.loads(open(timefile,'r').read())
vidfile = os.path.join(self.trial_dir, self.trial_name+'-cam.avi')
if not os.path.exists(vidfile):
vidfile = os.path.join(self.trial_dir, self.trial_name+'-cam0.avi')
if not os.path.exists(vidfile):
raise Exception('Movie %s not found.'%vidfile)
self.mov = VideoCapture(vidfile)
self.results = {}
self.results['centers'] = []
self.results['centers_all'] = []
self.results['left'] = 0
self.results['right'] = 0
self.results['middle'] = 0
self.results['left_assumed'] = 0
self.results['right_assumed'] = 0
self.results['middle_assumed'] = 0
self.results['skipped'] = 0
self.results['heat'] = np.zeros(np.shape(self.background))
self.results['n_frames'] = 0
self.results['params'] = [self.diff_thresh, self.kernel, self.translation_max, self.resample]
self.results['params_key'] = ['diff_thresh','kernel','translation_max','resample']
self.path_l, self.path_r, self.path_c, self.rooms_mask, self.paths_ignore, self.last_center = self.get_pt_selections()
def __init__(self,src_id,size=(640,480),fps=None):
self.controls = None
self.cvId = src_id
self.name = "VideoCapture"
self.controls = None
###add cv videocapture capabilities
self.capture = VideoCapture(src_id)
self.set_size(size)
def grabImageFromUSB(cameraNumber=0):
'''Grabs a snapshot from the specified USB camera.
Returns bool, video frame decoded as a JPEG bytearray.
'''
from cv2 import VideoCapture, imencode
# initialize the camera
cam = VideoCapture(cameraNumber)
retVal, rawData = cam.read()
if not retVal:
# frame captured returns errors
return False, None
retVal, jpgData = imencode('.jpg', rawData)
if not retVal:
# image encode errors
return False, None
return retVal, bytearray(jpgData)
class FileCapture():
"""
simple file capture that can auto_rewind
"""
def __init__(self,src):
self.auto_rewind = True
self.controls = None #No UVC controls available with file capture
# we initialize the actual capture based on cv2.VideoCapture
self.cap = VideoCapture(src)
self._get_frame_ = self.cap.read
def get_size(self):
return self.cap.get(3),self.cap.get(4)
def set_fps(self):
pass
def get_fps(self):
return None
def read(self):
s, img =self._get_frame_()
if self.auto_rewind and not s:
self.rewind()
s, img = self._get_frame_()
return s,img
def get_frame(self):
s, img = self.read()
timestamp = time()
return Frame(timestamp,img)
def rewind(self):
self.cap.set(1,0) #seek to the beginning
def create_atb_bar(self,pos):
return 0,0
def kill_atb_bar(self):
pass
def close(self):
pass
class Window(object):
def __init__(self, title="Video Stream"):
''' Uses OpenCV 2.3.1 method of accessing camera '''
self.title = title
self.cap = VideoCapture(0)
self.prev = self.get_frame()
self.frame = self.get_frame()
namedWindow(title, 1)
def get_frame(self):
success, frame = self.cap.read()
return self.to_grayscale(frame) if success else False
def to_grayscale(self, frame):
return cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
def optical_flow(self): # FIXME
flow = CreateImage(GetSize(frame), 32, 2)
CalcOpticalFlowFarneback(self.prev, self.frame, flow, # takes 0.05s
pyr_scale=0.5, levels=3, winsize=15,
iterations=3, poly_n=5, poly_sigma=1.2, flags=0)
return flow
def disparity(self):
if self.left is None or self.right is None:
print "Capture left and right images using 'l' and 'r' keys before running disparity"
return None
hl, wl = self.left.shape
hr, wr = self.right.shape
disp_left = cv2.cv.CreateMat(hl, wl, cv2.cv.CV_16S)
disp_right = cv2.cv.CreateMat(hr, wr, cv2.cv.CV_16S)
state = cv2.cv.CreateStereoGCState(16,2)
# running the graph-cut algorithm
from cv2.cv import fromarray
cv2.cv.FindStereoCorrespondenceGC(fromarray(self.left), fromarray(self.right), disp_left, disp_right, state)
cv2.cv.Save( "left.png", disp_left) # save the map
cv2.cv.Save( "right.pgm", disp_right) # save the map
def mainloop(self):
while True:
self.prev = self.frame
self.frame = self.get_frame()
sift(self.frame) # takes ~0.14s!
imshow(self.title, self.frame)
k = waitKey(10)
if k == -1:
pass
elif chr(k) == 'l':
self.left = self.frame
elif chr(k) == 'r':
self.right = self.frame
elif chr(k) == 'd':
self.disparity()
elif k == 27:
break
def check_cameras_linux(self, num=MAX_CAMERAS):
"""Comprueba las cámaras disponibles.
:Param num: máximo número de cámaras a comprobar
:Keyword num: 99 por defecto, ya que en Linux es lo permitido
:Param num: int
:Return: lista de cámaras disponibles
:Rtype: list of Capture
"""
n = 0
while len(self.cameras) < num and n <= MAX_CAMERAS:
camera = VideoCapture(n)
retval, frame = camera.read()
if retval:
self.cameras.append(camera)
n += 1
if num != MAX_CAMERAS and len(self.cameras) != num:
print("Found %d of %d cameras. " % (len(self.cameras), num))
exit()
return len(self.cameras)
class Camera_Capture():
"""
VideoCapture without uvc control using cv2.VideoCapture
"""
def __init__(self,src_id,size=(640,480),fps=None):
self.controls = None
self.cvId = src_id
self.name = "VideoCapture"
self.controls = None
###add cv videocapture capabilities
self.capture = VideoCapture(src_id)
self.set_size(size)
def get_frame(self):
s, img = self.capture.read()
timestamp = time()
return Frame(timestamp,img)
def set_size(self,size):
width,height = size
self.capture.set(3, width)
self.capture.set(4, height)
def get_size(self):
return self.capture.get(3), self.capture.get(4)
def set_fps(self,fps):
self.capture.set(5,fps)
def get_fps(self):
return self.capture.get(5)
def create_atb_bar(self,pos):
size = 0,0
return size
def kill_atb_bar(self):
pass
def close(self):
pass
def video_loop(aframes_queue,person_queue):
vc = VideoCapture(0)
rval, frame = vc.read()
people = {}
colors = ((0,0,255),(255,255,0))
while True:
rval, frame = vc.read()
if frame is None:
c = waitKey(10)
continue
aframe = NP.asarray(frame[:,:])
im = Image.fromarray(frame)
draw = ImageDraw.Draw(im)
while not person_queue.empty():
name,rect,name_size = person_queue.get()
people[name] = {'rect' : rect, 'name_size' : name_size,
'time_found' : time.time()}
name_counter = 0
for name in people.keys():
if name_counter < 2:
draw_name(draw, people[name], name, name_counter, colors[name_counter])
name_counter += 1
if time.time()>people[name]['time_found']+2:
# stop displaying after 2 seconds
people.pop(name)
frame2 = NP.array(im)
imshow('frame',frame2)
if aframes_queue.empty():
aframes_queue.put(aframe)
c = waitKey(1)
if c == 27: # exit on ESC
break
vc.release()
destroyAllWindows()
class noUVCCapture():
"""
VideoCapture without uvc control
"""
def __init__(self, src,size=(640,480)):
self.controls = None
self.cvId = src
###add cv videocapture capabilities
self.cap = VideoCapture(src)
self.set_size(size)
self.read = self.cap.read
def set_size(self,size):
width,height = size
self.cap.set(3, width)
self.cap.set(4, height)
def get_size(self):
return self.cap.get(3),self.cap.get(4)
def read_RGB(self):
s,img = self.read()
if s:
cvtColor(img,COLOR_RGB2BGR,img)
return s,img
def read_HSV(self):
s,img = self.read()
if s:
cvtColor(img,COLOR_RGB2HSV,img)
return s,img
class CameraCapture(uvc.Camera):
"""
CameraCapture Class for Image grabbing and control
inherits from an OS specitic Camera that defines all uvc control methods
"""
def __init__(self, cam,size=(640,480)):
super(CameraCapture, self).__init__(cam)
###add cv videocapture capabilities
self.cap = VideoCapture(self.cvId)
self.set_size(size)
self.read = self.cap.read
def set_size(self,size):
width,height = size
self.cap.set(3, width)
self.cap.set(4, height)
def get_size(self):
return self.cap.get(3),self.cap.get(4)
def read_RGB(self):
s,img = self.read()
if s:
cvtColor(img,COLOR_RGB2BGR,img)
return s,img
def read_HSV(self):
s,img = self.read()
if s:
cvtColor(img,COLOR_RGB2HSV,img)
return s,img
def __init__(self, cam, size=(640, 480), fps=30):
self.src_id = cam.src_id
self.uId = cam.uId
self.name = cam.name
self.controls = Controls(self.uId)
try:
self.controls["UVCC_REQ_FOCUS_AUTO"].set_val(0)
except KeyError:
pass
self.capture = VideoCapture(self.src_id)
self.set_size(size)
def __init__(self,src):
self.auto_rewind = True
self.controls = None #No UVC controls available with file capture
# we initialize the actual capture based on cv2.VideoCapture
self.cap = VideoCapture(src)
timestamps_loc = os.path.join(src.rsplit(os.path.sep,1)[0],'eye_timestamps.npy')
logger.info("trying to load timestamps with video at: %s"%timestamps_loc)
try:
self.timestamps = np.load(timestamps_loc).tolist()
logger.info("loaded %s timestamps"%len(self.timestamps))
except:
logger.info("did not find timestamps")
self.timestamps = None
self._get_frame_ = self.cap.read
class FileCapture():
"""
simple file capture that can auto_rewind
"""
def __init__(self,src):
self.auto_rewind = True
self.controls = None #No UVC controls available with file capture
# we initialize the actual capture based on cv2.VideoCapture
self.cap = VideoCapture(src)
self._get_frame_ = self.cap.read
def get_size(self):
return self.cap.get(3),self.cap.get(4)
def read(self):
s, img =self._get_frame_()
if self.auto_rewind and not s:
self.rewind()
s, img = self._get_frame_()
return s,img
def read_RGB(self):
s,img = self.read()
if s:
cvtColor(img,COLOR_RGB2BGR,img)
return s,img
def read_HSV(self):
s,img = self.read()
if s:
cvtColor(img,COLOR_RGB2HSV,img)
return s,img
def rewind(self):
self.cap.set(1,0) #seek to the beginning
def __init__(self,src_id,size=(640,480),fps=None,timebase=None):
self.controls = None
self.cvId = src_id
self.name = "VideoCapture"
self.controls = None
###add cv videocapture capabilities
self.capture = VideoCapture(src_id)
self.set_size(size)
if timebase == None:
logger.debug("Capture will run with default system timebase")
self.timebase = c_double(0)
elif isinstance(timebase,c_double):
logger.debug("Capture will run with app wide adjustable timebase")
self.timebase = timebase
else:
logger.error("Invalid timebase variable type. Will use default system timebase")
self.timebase = c_double(0)
def re_init(self, cam, size=(640, 480), fps=30):
self.src_id = cam.src_id
self.uId = cam.uId
self.name = cam.name
self.controls = Controls(self.uId)
try:
self.controls["UVCC_REQ_FOCUS_AUTO"].set_val(0)
except KeyError:
pass
self.capture = VideoCapture(self.src_id)
self.set_size(size)
# recreate the bar with new values
bar_pos = self.bar._get_position()
self.bar.destroy()
self.create_atb_bar(bar_pos)
def __init__(self):
try:
self.camera = VideoCapture(CAMERA_INDEX)
self.camera.set(3,CAMERA_WIDTH)
self.camera.set(4,CAMERA_HEIGHT)
message = 'Success.'
except Exception:
message = 'Failure.'
print('[Setting up Camera]...' + message)
try:
self.arduino = serial.Serial(DEVICE, BAUD)
message = 'Success.'
except Exception:
message = 'Failure.'
print('[Setting up Controller]...' + message)
self.reset_worker()
def __init__(self, cam,size=(640,480),fps=30):
self.src_id = cam.src_id
self.uId = cam.uId
self.name = cam.name
self.controls = Controls(self.uId)
try:
self.controls['UVCC_REQ_FOCUS_AUTO'].set_val(0)
except KeyError:
pass
if '6000' in self.name and False: #on mac we dont have enough controls to use this right.
logger.info("adjusting exposure for HD-6000 camera")
try:
self.controls['UVCC_REQ_EXPOSURE_AUTOMODE'].set_val(1)
self.controls['UVCC_REQ_EXPOSURE_ABS'].set_val(156)
except KeyError:
pass
self.capture = VideoCapture(self.src_id)
self.set_size(size)
class LKTrack:
_lk_params = dict(winSize=(11, 11),
maxLevel=2,
criteria=(TERM_CRITERIA_EPS | TERM_CRITERIA_COUNT, 10,
.003))
_feature_params = dict(maxCorners=500, qualityLevel=.01, minDistance=10)
def __init__(self):
self._root = Tk()
self._flag_run = False
self._frames = Label(self._root, text='frames: ')
self._in_p = Label(self._root, text='in: 0')
self._file_name = Label(self._root, text='file name: ')
self._button_file = Button(self._root,
text='open file',
command=self._open_run_muvie)
self._button_cum = Button(self._root,
text='run on cum',
command=self._open_run_cum)
self._button_file.grid(row=0, column=0)
self._button_cum.grid(row=0, column=1)
self._file_name.grid(row=0, column=2)
self._frames.grid(row=0, column=3)
self._in_p.grid(row=0, column=4)
self._build_online_rectangle = []
self._cordinate_squade = self._load_squade()
def _get_lmain(self):
self._lmain = Label(self._root)
self._lmain.bind('<Motion>', self._make_squade)
self._lmain.grid(row=1, column=0, columnspan=5)
def _open_run_muvie(self):
open_file = filedialog.askopenfilename()
if open_file:
self._cam = VideoCapture(open_file)
self._prepare_for_word()
self._file_name['text'] = 'file name: ' + open_file.split('/')[-1]
def _prepare_for_word(self):
if self._flag_run:
self._lmain.destroy()
self._flag_run = True
self._get_lmain()
self._current_frame = self._people_in = 0
self._track = []
self._detect_points()
def _open_run_cum(self):
self._cam = VideoCapture(0)
self._prepare_for_word()
self._file_name['text'] = 'run on cam'
def _detect_points(self):
_, self._image = self._cam.read()
self._gray = cvtColor(self._image, COLOR_BGR2GRAY)
features = goodFeaturesToTrack(
self._gray[self._cordinate_squade[2]:self._cordinate_squade[3],
self._cordinate_squade[0]:self._cordinate_squade[1]],
**self._feature_params)
try:
features += np.array(
[[self._cordinate_squade[0], self._cordinate_squade[2]]],
dtype=np.float32)
for x, y in features.reshape((-1, 2)):
self._track.append([(x, y)])
except:
pass
self._prev_gray = self._gray
def _track_points(self):
_, self._image = self._cam.read()
self._gray = cvtColor(self._image, COLOR_BGR2GRAY)
tmp = np.float32([tp[-1] for tp in self._track]).reshape(-1, 1, 2)
features, status, _ = calcOpticalFlowPyrLK(self._prev_gray, self._gray,
tmp, None,
**self._lk_params)
features = [p for (st, p) in zip(status, features) if st]
features = np.array(features).reshape((-1, 2))
ndx = [i for (i, st) in enumerate(status) if not st]
ndx.reverse()
for i in ndx:
self._track.pop(i)
for i, f in enumerate(features):
self._track[i].append(tuple(f))
self._track = [i[-100:] for i in self._track]
self._prev_gray = self._gray
def _point_is_move(self, point_track):
if (abs(point_track[0][0] - point_track[-1][0]) >
30) and (point_track[-1][0] > self._cordinate_squade[1]):
return True
return False
def _centroid(self):
points = [p for p in self._track if self._point_is_move(p)]
if len(points) > 10:
mean_x = int(sum([p[-1][0] for p in points]) // len(points))
mean_y = int(sum([p[-1][1] for p in points]) // len(points))
circle(self._image, (mean_x, mean_y), 10, (200, 0, 0), -1)
if mean_x > self._cordinate_squade[1] + 90:
for i in points:
self._track.remove(i)
self._people_in += 1
self._in_p['text'] = f'in {self._people_in}'
self._save_img()
def _draw_points(self):
for index, (x, y) in enumerate((i[-1] for i in self._track)):
color = (255, 0,
0) if self._point_is_move(self._track[index]) else (0,
255,
0)
circle(self._image, (int(x), int(y)), 3, color, -1)
def _draw_rectangle(self):
rectangle(self._image,
(self._cordinate_squade[0], self._cordinate_squade[-1]),
(self._cordinate_squade[1], self._cordinate_squade[2]),
(0, 0, 255))
def _update_widget(self):
self._current_frame += 1
self._frames['text'] = f"frames: {self._current_frame}"
def _update_window(self):
img = Image.fromarray(self._image)
imgtk = ImageTk.PhotoImage(image=img)
self._lmain.imgtk = imgtk
self._lmain.configure(image=imgtk)
def _save_img(self):
imsave(str(self._current_frame) + '.jpg', self._image)
def _save_squade(self):
with open('cordinate_squade.txt', 'w') as file:
file.write(str(self._cordinate_squade))
def _load_squade(self):
with open('cordinate_squade.txt') as file:
return eval(file.read())
def _make_squade(self, event):
if event.state == 264:
self._build_online_rectangle.append((event.x, event.y))
elif self._build_online_rectangle:
self._cordinate_squade.clear()
self._cordinate_squade.append(
min((self._build_online_rectangle[0][0],
self._build_online_rectangle[-1][0])))
self._cordinate_squade.append(
max((self._build_online_rectangle[0][0],
self._build_online_rectangle[-1][0])))
self._cordinate_squade.append(
min((self._build_online_rectangle[0][1],
self._build_online_rectangle[-1][1])))
self._cordinate_squade.append(
max((self._build_online_rectangle[0][1],
self._build_online_rectangle[-1][1])))
self._build_online_rectangle.clear()
self._save_squade()
def _del_static_points(self):
falg = True
for point in self._track:
if (point[-1][0] > self._cordinate_squade[1]
) and not self._point_is_move(point):
self._track.remove(point)
def run(self):
if self._flag_run:
if len(self._track) < 200:
if not (self._current_frame % 200) and (
not [i
for i in self._track if self._point_is_move(i)]):
self._track.clear()
self._detect_points()
self._track_points()
self._update_widget()
self._draw_points()
self._del_static_points()
self._centroid()
self._draw_rectangle()
self._update_window()
self._root.after(1, self.run)
return self._root
def get_fps(video: VideoCapture) -> float:
return video.get(cv2.CAP_PROP_FPS)
def _open_run_cum(self):
self._cam = VideoCapture(0)
self._prepare_for_word()
self._file_name['text'] = 'run on cam'
class Video_reader:
def __init__(self, filename):
self.curr_frame = 0 # BE careful !!! IT IS NOT ALWAYS UPDATED !!!!
self.frames_count = 0
print("Initializing video reader.")
self.filename = filename
print("Opening video file.")
self.video = None
try:
self.video = VideoCapture(filename)
except:
print("Videofile failed to load.")
if self.is_open():
self.frames_count = self.video.get(7)
print("Video " + filename + " is opened.")
def read_frame(self, frame=-1):
if frame is None:
return None
if frame <= -1:
ret, frame = self.video.read()
self.curr_frame += 1
if ret:
return frame
try:
self.video.set(1, int(frame))
except:
return None
self.curr_frame = frame
ret, frame = self.video.read()
if ret:
return frame
def read_frame_at_time(self, time):
self.video.set(0, time)
ret, frame = self.video.read()
if ret:
return frame
def set_position_in_ms(self, time):
if self.video is not None:
self.video.set(0, time)
def set_position_frame(self, frame):
if self.video is not None:
if frame < 0:
self.video.set(1, 0)
elif frame > self.frames_count:
self.video.set(1, self.frames_count - 1)
else:
self.video.set(1, frame)
# self.curr_frame = frame
def get_position_in_ms(self):
if self.video is not None:
return self.video.get(0)
def get_position_frames(self):
# self.curr_frame = self.video.get(1)
if self.video is not None:
return self.video.get(1)
def get_position_ratio(self):
if self.video is not None:
return self.video.get(2)
def get_frame_width(self):
if self.video is not None:
return self.video.get(3)
def get_frame_height(self):
if self.video is not None:
return self.video.get(4)
def get_fps(self):
if self.video is not None:
return self.video.get(5)
def get_fourcc(self):
if self.video is not None:
return self.video.get(6)
def get_frames_count(self):
return self.frames_count
def get_filename(self):
return self.filename
def is_open(self):
if self.video is not None:
return self.video.isOpened()
else:
return False
return ascii_img
printIntro()
pixel_size = int(input('Enter the size of pixel: '))
is_update_display = False
if (input('Enable console clear [y/n] ? ') == 'y'):
is_update_display = True
else:
is_update_display = False
print('Please wait...')
cap = VideoCapture(0)
while True:
_, img = cap.read()
h, _, _ = img.shape
gray_img = cvtColor(img, COLOR_RGB2GRAY)
gray_img = flip(gray_img,1)
output = makeAscii(gray_img, h, pixel_size)
output = [''.join(output[i]) for i in range(h // pixel_size)]
output = '\n'.join(output)
if (is_update_display):
clear()
else:
print('\n' * 150)
print(output)
def get_current_frame_id(video_cap: cv2.VideoCapture) -> int:
return int(video_cap.get(cv2.CAP_PROP_POS_FRAMES))
def get_frame_count(video_cap: cv2.VideoCapture) -> int:
# NOT always accurate, see:
# https://stackoverflow.com/questions/31472155/python-opencv-cv2-cv-cv-cap-prop-frame-count-get-wrong-numbers
return int(video_cap.get(cv2.CAP_PROP_FRAME_COUNT))
def get_video_capture_specs(cap: cv.VideoCapture) -> do.VideoSpecs:
if not isinstance(cap, cv.VideoCapture):
raise TypeError
return do.VideoSpecs(width=cap.get(cv.CAP_PROP_FRAME_WIDTH),
height=cap.get(cv.CAP_PROP_FRAME_HEIGHT),
fps=cap.get(cv.CAP_PROP_FPS))
else:
filename = filename \
+ '_Location:' + row + col + '_Heading:' + heading + '_Ship:' + ship
print('SHIP: ' + ship)
print('HEADING: ' + heading)
print('LOCATION: ' + row + col)
# Assign the approriate directory to the output file
if np.random.rand() < fraction_train:
filename = dataset_dirs[0] + filename
else:
filename = dataset_dirs[1] + filename
# Initialize the camera capture object with the cv2.VideoCapture class.
# All it needs is the index to a camera port.
camera = VideoCapture(1)
key = raw_input('When ready, press Enter to capture image.')
# Ramp the camera - these frames will be discarded and are only used
# to allow v4l2 to adjust light levels, if necessary
print('Ramping camera...')
for i in xrange(ramp_frames):
return_val, image = camera.read()
# Take the actual image we want to keep
print('Capturing actual image...')
return_val, image = camera.read()
# Save the image in JPG format
filename += '.jpg'
print('Writing image: ' + filename)
THRESHOLD = 1e-4
audio = pyaudio.PyAudio()
# start Recording
stream = audio.open(format=FORMAT,
channels=CHANNELS,
rate=RATE,
input=True,
frames_per_buffer=CHUNK)
while True:
cur_chunk = np.abs(np.fromstring(stream.read(CHUNK)))
convolved = np.convolve(KERNEL, cur_chunk)
if np.max(convolved) > THRESHOLD:
vc = VideoCapture(0)
if vc.isOpened():
rval, frame = vc.read()
if rval:
imwrite('/tmp/.out-clapchat.png', frame)
vc.release()
break
# stop Recording
stream.stop_stream()
stream.close()
audio.terminate()
toaddr = input("Enter recipient email addresses separated by spaces: ").split()
fromaddr = "*****@*****.**"
def __init__(self, configuration, names, type='video', convert_to_grayscale=False,
progress_signal=None):
"""
Initialize the Frame object, and read all images. Images can be stored in a video file or
as single images in a directory.
:param configuration: Configuration object with parameters
:param names: In case "video": name of the video file. In case "image": list of names for
all images.
:param type: Either "video" or "image".
:param convert_to_grayscale: If "True", convert frames to grayscale if they are RGB.
:param progress_signal: Either None (no progress signalling), or a signal with the signature
(str, int) with the current activity (str) and the progress in
percent (int).
"""
self.configuration = configuration
self.progress_signal = progress_signal
if type == 'image':
# Use scipy.misc to read in image files. If "convert_to_grayscale" is True, convert
# pixel values to 32bit floats.
self.number = len(names)
self.signal_step_size = max(int(self.number / 10), 1)
if convert_to_grayscale:
self.frames_original = [misc.imread(path, mode='F') for path in names]
else:
self.frames_original = []
for frame_index, path in enumerate(names):
# After every "signal_step_size"th frame, send a progress signal to the main GUI.
if self.progress_signal is not None and frame_index%self.signal_step_size == 0:
self.progress_signal.emit("Read all frames",
int((frame_index / self.number) * 100.))
# Read the next frame.
frame = imread(path, -1)
self.frames_original.append(frame)
if self.progress_signal is not None:
self.progress_signal.emit("Read all frames", 100)
self.shape = self.frames_original[0].shape
dt0 = self.frames_original[0].dtype
# Test if all images have the same shape, color type and depth.
# If not, raise an exception.
for image in self.frames_original:
if image.shape != self.shape:
raise ShapeError("Images have different size")
elif len(self.shape) != len(image.shape):
raise ShapeError("Mixing grayscale and color images not supported")
if image.dtype != dt0:
raise TypeError("Images have different type")
elif type == 'video':
# In case "video", use OpenCV to capture frames from video file. Revert the implicit
# conversion from RGB to BGR in OpenCV input.
cap = VideoCapture(names)
self.number = int(cap.get(CAP_PROP_FRAME_COUNT))
self.frames_original = []
self.signal_step_size = max(int(self.number / 10), 1)
for frame_index in range(self.number):
# After every "signal_step_size"th frame, send a progress signal to the main GUI.
if self.progress_signal is not None and frame_index%self.signal_step_size == 0:
self.progress_signal.emit("Read all frames", int((frame_index/self.number)*100.))
# Read the next frame.
ret, frame = cap.read()
if ret:
if convert_to_grayscale:
self.frames_original.append(cvtColor(frame, COLOR_BGR2GRAY))
else:
self.frames_original.append(cvtColor(frame, COLOR_BGR2RGB))
else:
raise IOError("Error in reading video frame")
cap.release()
if self.progress_signal is not None:
self.progress_signal.emit("Read all frames", 100)
self.shape = self.frames_original[0].shape
dt0 = self.frames_original[0].dtype
else:
raise TypeError("Image type " + type + " not supported")
# Monochrome images are stored as 2D arrays, color images as 3D.
if len(self.shape) == 2:
self.color = False
elif len(self.shape) == 3:
self.color = True
else:
raise ShapeError("Image shape not supported")
# Set the depth value of all images to either 16 or 8 bits.
if dt0 == 'uint16':
self.depth = 16
elif dt0 == 'uint8':
self.depth = 8
else:
raise TypeError("Frame type " + str(dt0) + " not supported")
# Initialize lists of monochrome frames (with and without Gaussian blur) and their
# Laplacians.
self.frames_monochrome = None
self.frames_monochrome_blurred = None
self.frames_monochrome_blurred_laplacian = None
self.used_alignment_points = None
def capture_images_continually(capture: cv2.VideoCapture, model, classes, img_size, device):
global outputFrame, LINE_COORD_COLOR, DETECT_TRAFFIC_LIGHT, TRAFFIC_LIGHT_RECT
violations = {}
tracked_paths = {}
track = sort.Sort()
i = 0
while True:
i += 1
ret, frame = capture.read()
t0 = 0
if PRINT_FRAME_DURATION:
t0 = time.time()
if not ret:
print('no video')
capture.set(cv2.CAP_PROP_POS_FRAMES, 0)
continue
with torch.no_grad():
boxes = detect.detect(model, frame, img_size, device=device)
boxes_cars = boxes[np.isin(boxes[:, 5], [2, 3, 5, 7])] # take car, motobuke, bus, truck
if DETECT_TRAFFIC_LIGHT:
DETECT_TRAFFIC_LIGHT = False
traffic_lights = boxes[np.isin(boxes[:, 5], [9])]
traffic_lights_coords = traffic_lights[:, :4]
lights = traffic_color(frame, traffic_lights_coords)
for color, light in lights:
if color != NO_COLOR:
TRAFFIC_LIGHT_RECT = (light[0], light[1]), (light[2], light[3])
if DETECT_TRAFFIC_LIGHT_COLOR:
nptraffic_coords = np.reshape(np.array([*TRAFFIC_LIGHT_RECT[0], *TRAFFIC_LIGHT_RECT[1]]), (1, 4))
lights = traffic_color(frame, nptraffic_coords)
detected_tr_lights = Counter(
list(filter(lambda x: x != NO_COLOR, map(lambda x: x[0], lights)))).most_common()
if len(detected_tr_lights) > 0:
LINE_COORD_COLOR = detected_tr_lights[0][0]
# draw traffic light
cv2.rectangle(frame, TRAFFIC_LIGHT_RECT[0], TRAFFIC_LIGHT_RECT[1], (255, 215, 0), 2)
cv2.putText(frame, LINE_COORD_COLOR, (int(TRAFFIC_LIGHT_RECT[0][0]), int(TRAFFIC_LIGHT_RECT[0][1] - 10)),
cv2.FONT_HERSHEY_SIMPLEX, 0.9,
(36, 255, 12), 2)
boxes_no_class = boxes_cars[:, :-1]
if DRAW_DETECTION_BOXES:
for x0, y0, x1, y1, confidence in boxes_no_class:
x0, y0, x1, y1 = map(int, [x0, y0, x1, y1])
cv2.rectangle(frame, (x0, y0), (x1, y1), (0, 255, 255), 2)
tracked_objects = track.update(boxes_no_class)
# process tracking boxes
for x0, y0, x1, y1, obj_id in tracked_objects:
x0, y0, x1, y1 = map(int, [x0, y0, x1, y1])
if obj_id not in tracked_paths:
tracked_paths[obj_id] = []
path = tracked_paths[obj_id]
path.append((int((x0 + x1) / 2.), int((y0 + y1) / 2.), time.time()))
if line_intersection(LINE_COORD, path) and LINE_COORD_COLOR == RED_OR_YELLOW:
if obj_id not in violations:
print("violation!")
# TODO: store something useful here, otherwise it's just a hashset
violations[obj_id] = None
if obj_id in violations:
rect_rgb = (0, 0, 215)
line_rgb = (0, 0, 215)
else:
rect_rgb = (0, 215, 0)
line_rgb = (0, 255, 0)
# draw tracking box
if DRAW_TRACKING_BOXES:
cv2.rectangle(frame, (x0, y0), (x1, y1), rect_rgb, 2)
for i in range(len(path) - 1):
cv2.line(frame, tuple(path[i][:2]), tuple(path[i + 1][:2]), line_rgb, 2)
cv2.line(frame, LINE_COORD[0], LINE_COORD[1], (255, 0, 0), 2)
if PRINT_FRAME_DURATION:
print(f"frame time: {time.time() - t0:.2}")
outputFrame = frame
# clean old paths (older than 30 seconds)
if i % 1000 == 0:
tracked_paths = {k: v for k, v in tracked_paths.items() if len(v) != 0}
for key, val in tracked_paths.items():
val[:] = [[*a, time_created] for *a, time_created in val if time.time() - time_created < 30]
def get_frame_size(video: VideoCapture) -> Tuple[float, float]:
return (
video.get(cv2.CAP_PROP_FRAME_HEIGHT),
video.get(cv2.CAP_PROP_FRAME_WIDTH)
)
def get_frame_count(video: VideoCapture) -> float:
return video.get(cv2.CAP_PROP_FRAME_COUNT)
def init(self):
self.cap = VideoCapture(
"test_videos/ntb/head_office/Cash_Counter_1-1.dav")
self.total_input_frames = 0
import cv2
import numpy as np
from cv2 import VideoCapture
def callback(x):pass
cap = VideoCapture(0)
cv2.namedWindow('image')
ilowH = 0
ihighH = 179
ilowS = 0
ihighS = 255
ilowV = 0
ihighV = 255
cv2.createTrackbar('lowH','image',ilowH,179,callback)
cv2.createTrackbar('highH','image',ihighH,179,callback)
cv2.createTrackbar('lowS','image',ilowS,255,callback)
cv2.createTrackbar('highS','image',ihighS,255,callback)
cv2.createTrackbar('lowV','image',ilowV,255,callback)
cv2.createTrackbar('highV','image',ihighV,255,callback)
detector = cv2.SimpleBlobDetector_create()
while True:
ret, frame = cap.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
ilowH = cv2.getTrackbarPos('lowH', 'image')
ihighH = cv2.getTrackbarPos('highH', 'image')
ilowS = cv2.getTrackbarPos('lowS', 'image')
ihighS = cv2.getTrackbarPos('highS', 'image')
ilowV = cv2.getTrackbarPos('lowV', 'image')
ihighV = cv2.getTrackbarPos('highV', 'image')
hsv_min = np.array([ilowH, ilowS, ilowV])
hsv_max = np.array([ihighH, ihighS, ihighV])
def video_jump(video_cap: cv2.VideoCapture, frame_id: int):
video_cap.set(cv2.CAP_PROP_POS_FRAMES, frame_id - 1)
futures.append(
executor.submit(ocr_function, sudoku[i, j], i, j))
for f in futures:
result, i, j = f.result()
sudoku_pf[i, j] = int(result)
print(sudoku_pf)
stop = time()
exec_time = stop - start
return sudoku_pf, True, exec_time
input = 0 if len(argv) == 1 else argv[1]
cap = VideoCapture(input) # Use video/webcam as input source
#image = imread("sudoku3.jpg") # Use image as input source
ret_val = True
sudoku = None
solved = False
# If True, ocr.space API is used, if False, Tesseract OCR is used
use_api = False
correct, incorrect = [], []
intersections, h_lines, v_lines = [], [], []
# while-loop keeps reading images from the source until ret_val is false,
# which means no image has been retrieved from the source
while ret_val:
if sudoku is None:
def get_current_frame_time(video_cap: cv2.VideoCapture) -> float:
return video_cap.get(cv2.CAP_PROP_POS_MSEC) / 1000
import cv2 as cv
from cv2 import VideoCapture
import numpy as np
# checking installation status & version of module
print(f'Version number is {cv.__version__} ')
# pass int 0 for first camera on our device
cap = VideoCapture('test/shesgone.mp4')
program_masih_berjalan = True
while program_masih_berjalan:
if (cap.isOpened() == False):
print('Error opening camera')
break
elif (cap.isOpened() == True):
# frame by frame capture
ret, frame = cap.read()
# if frame is read correctly ret is True
if not ret:
print("Can't receive frame (stream end?). Exiting ...")
break
# Our operations on the frame come here
gray = cv.cvtColor(frame, cv.COLOR_BGR2GRAY)
# Display the resulting frame
cv.imshow('frame', gray)
if cv.waitKey(1) == ord('q'):
cap.release()
cv.destroyAllWindows()
break
def get_frame_size(video_cap: cv2.VideoCapture) -> typing.Tuple[int, int]:
""" return size of frame: (width, height) """
h = video_cap.get(cv2.CAP_PROP_FRAME_HEIGHT)
w = video_cap.get(cv2.CAP_PROP_FRAME_WIDTH)
return int(w), int(h)
# Load the labels from COCO.
with open(labels_file, 'r') as lines:
coco_labels = lines.readlines()[0].split(',')
# Import the DarkNet model, and load the pre-trained weights.
net = DarkNet(cfg_file, weights_file)
# Get the colors tha will be used to create the bounding boxes.
np.random.seed(42)
colors = np.random.randint(0, 255, size=(len(coco_labels), 3))
# Part 1: Video processing.
if img_or_vid == 'vid':
# Use OpenCV to load the video and obtain the number of frames.
video = VC(input_file)
number_of_frames = int(video.get(n_frames))
fps = video.get(n_fps)
# Create a new video.
_, first_frame = video.read()
video_shape = first_frame.shape
new_video = CW(output_file,
cv2.VideoWriter_fourcc('M', 'J', 'P', 'G'),
mt.floor(fps), (video_shape[1], video_shape[0]),
isColor=True)
# Process the video.
classes = []
count = 0
n = 1 # Change this value if you want to process non-sequential frame.
def calibrate_affine(
src_cam: cv.VideoCapture,
dst_cam: cv.VideoCapture,
n_images=10,
chessboard_size=(6, 9),
size=(640, 480),
):
src_points = []
dst_points = []
n_captured_images = 0
while n_captured_images < n_images:
# read images from both cameras
is_ok, src_image = src_cam.read()
if not is_ok:
continue
is_ok, dst_image = dst_cam.read()
if not is_ok:
continue
# resize
src_resized = utils.resize(src_image, size)
dst_resized = utils.resize(dst_image, size)
# get BGR and Gray
src_bgr, src_gray = utils.to_bgr_gray(src_resized)
dst_bgr, dst_gray = utils.to_bgr_gray(dst_resized)
# get corners
src_corners_ok, src_corners = utils.get_chessboard_corners(
src_gray, chessboard_size)
dst_corners_ok, dst_corners = utils.get_chessboard_corners(
dst_gray, chessboard_size)
# draw
cv.drawChessboardCorners(src_bgr, chessboard_size, src_corners,
src_corners_ok)
cv.drawChessboardCorners(dst_bgr, chessboard_size, dst_corners,
dst_corners_ok)
cv.putText(
src_bgr,
f"{n_captured_images}/{n_images}",
(0, src_bgr.shape[0]),
cv.FONT_HERSHEY_SIMPLEX,
2,
(255, 0, 0),
)
cv.putText(
dst_bgr,
f"{n_captured_images}/{n_images}",
(0, dst_bgr.shape[0]),
cv.FONT_HERSHEY_SIMPLEX,
2,
(255, 0, 0),
)
cv.imshow('src', src_bgr)
cv.imshow('dst', dst_bgr)
# save if ok
both_ok = src_corners_ok and dst_corners_ok
wait_time = 10
if both_ok:
wait_time = -1
key = cv.waitKey(wait_time)
if key == 32 and both_ok:
n_captured_images += 1
src_points.append(src_corners)
dst_points.append(dst_corners)
cv.destroyAllWindows()
# calc affine transform
np_src_points = np.reshape(src_points, (-1, 2))
np_dst_points = np.reshape(dst_points, (-1, 2))
trans_mat, inliers = cv.estimateAffinePartial2D(np_src_points,
np_dst_points,
method=cv.RANSAC)
return trans_mat
def get_video_capture(self, video_path: Path) -> VideoCapture:
return VideoCapture(str(video_path))
# Comprueba el tiempo que tarda en predecir la imagen sin humo
before = datetime.now()
# Predice la imagen sin humo
predictionNeutral = model.predict(imageNeutral, batch_size=1)
after = datetime.now()
print('Prediction time neutral image: ' +
str((after - before).total_seconds()))
# Crea las máscaras de cada imagen
maskSmoke = threshold((predictionSmoke[0][..., 0] * 255.).astype('uint8'), 127,
255, 0)[1].astype('uint8')
maskNeutral = threshold((predictionNeutral[0][..., 0] * 255.).astype('uint8'),
127, 255, 0)[1].astype('uint8')
# Guarda las imágenes redimensionadas y las máscaras predecidas
imwrite('/root/Smoke/Test/smokeMask.png', maskSmoke)
imwrite('/root/Smoke/Test/smokeAug.png', cvtColor(imageSmokeAug, 4))
imwrite('/root/Smoke/Test/neutralMask.png', maskNeutral)
imwrite('/root/Smoke/Test/neutralAug.png', cvtColor(imageNeutralAug, 4))
# Imprime si la predicción ha fallado o ha sido exitosa
if sum(sum(maskSmoke)) <= 0 or sum(sum(maskNeutral)) > 0:
print('Prediction FAILED!')
else:
print('Prediction SUCCESSFUL!')
# Realiza una captura de cada cámara
for i in range(1):
imwrite('/root/Smoke/TestCam/' + 'imageCamera' + str(i) + '.png',
VideoCapture(i).read()[1])
VideoCapture(i).release()
# Devuelve el tiempo total que le ha llevado la prueba
afterAll = datetime.now()
print('Time: ' + str((afterAll - beforeAll).total_seconds()))
def __init__(self, window):
super(GamePage, self).__init__(window)
# ------------------------- 游戏是否结束 -------------------------#
self.hasGameOver = False
# ------------------------- 左右分数 -------------------------#
self.leftScore = 0
self.rightScore = 0
# ------------------------- 游戏模式 -------------------------#
# 检测人脸的个数
if BasePage.mode == MODE.SINGLE:
self.faceCount = 1
elif BasePage.mode == MODE.DOUBLE:
self.faceCount = 2
# ------------------------- 暂停 -------------------------#
# 是否暂停
self.isPause = False
# 暂停文字
# 标题文本
self.pauseSurface = self.middleFont.render('Pause', True, WHITE)
# 获取文本的矩形
self.pauseRect = self.pauseSurface.get_rect()
# 设置矩形中心点
self.pauseRect.center = (WINDOW_WIDTH / 2, WINDOW_HEIGHT / 2)
# ------------------------- 检测的人脸方框数据 -------------------------#
self.faceRects = []
# ------------------------- 存放球拍数据 -------------------------#
self.racketSurfaces = []
# ------------------------- 球拍 -------------------------#
# 乒乓球拍
self.racketSurface = image.load('image/pai.png')
# 球拍的宽度和高度
self.racketW = self.racketSurface.get_width()
self.racketH = self.racketSurface.get_height()
# ------------------------- 小球位置 -------------------------#
# 小球位置
self.pos_x = 300 # 矩形的X轴坐标
self.pos_y = 250 # 矩形的Y轴坐标
# 小球x和y轴分速度
self.vel_x = 24 # X轴分速度
self.vel_y = 12 # Y轴分速度
# ------------------------- 球 -------------------------#
# 乒乓球拍
ballSurface = image.load('image/ball.png')
# 球拍的宽度和高度
w = ballSurface.get_width()
h = ballSurface.get_height()
self.ballSurface = transform.scale(ballSurface, (w // 20, h // 20))
# 球矩形
self.ballRect = self.ballSurface.get_rect()
# 球的宽度和高度
self.ballW = self.ballRect.width
self.ballH = self.ballRect.height
# 设置矩形中心点
self.ballRect.topleft = (self.pos_x, self.pos_y)
# tensorflow检测类
self.tfDetector = TensoflowFaceDector(PATH_TO_CKPT)
# ---------------------- 读取配置获取摄像头id ---------------------- #
f = open('config/config.txt')
lines = f.readlines()
for line in lines:
if line.startswith('camera_index'):
CAMERA_INDEX = int(line[line.index(':') + 1:])
elif line.startswith('game_over_score'):
self.game_over_score = int(line[line.index(':') + 1:])
print('---------------摄像头id:', CAMERA_INDEX)
# 摄像头
self.camera = VideoCapture(CAMERA_INDEX)
self.camera.set(3, WINDOW_WIDTH)
self.camera.set(4, WINDOW_HEIGHT)
def approximate(self, source_path, algorithm, num_points):
# Check to ensure that a file exists at the specified path
if (not video_exists(source_path)):
logger.error("Failed to find a video file at the specified path.")
return
# Initialize the views directory
VIEWS_DIRECTORY = "views"
initialize_views_directory(VIEWS_DIRECTORY)
# Setting up video reader
video = VideoCapture(source_path)
# Find the dimensions of the video to define the video cube
num_frames, video_width, video_height, fps = get_video_parameters(
video)
vc = VideoCube(num_frames, video_height, video_width)
# Setting up video writer
fourcc = VideoWriter_fourcc(*'mp4v')
video_out = VideoWriter(VIEWS_DIRECTORY + "/output.mp4", fourcc, fps,
(video_height, video_width))
video_out_lines = VideoWriter(
VIEWS_DIRECTORY + "/output_view_lines.mp4", fourcc, fps,
(video_height, video_width))
video_out_corners = VideoWriter(
VIEWS_DIRECTORY + "/output_view_corners.mp4", fourcc, fps,
(video_height, video_width))
video_out_keypoints = VideoWriter(
VIEWS_DIRECTORY + "/output_view_keypoints.mp4", fourcc, fps,
(video_height, video_width))
# Initialize the video cube according to the point initialization algorithm
# Exit if points failed to be placed
if not self._initialize_point_placer(algorithm, vc, num_points, video):
logger.error("Point placement failed to initialize.")
return
# Tetrahedralize the video cube
vc.tetrahedralize()
# Loop through the
frame_number = 0
while video.isOpened():
# Read a frame of the video
frames_remain, frame = video.read()
# Stop reading if we reach the end of the video
if not frames_remain:
break
# Slice the video cube at this frame and create a low poly frame
frame_lp, frame_lp_lines, frame_lp_corners, lp_frame_keypoints = vc.slice_cube(
frame, frame_number)
frame_number += 1
# Write the low poly frame
video_out.write(frame_lp)
video_out_lines.write(frame_lp_lines)
video_out_corners.write(frame_lp_corners)
video_out_keypoints.write(lp_frame_keypoints)
# Release video reader and writer
video_out.release()
video_out_lines.release()
video_out_corners.release()
video_out_keypoints.release()
video.release()
def _open_run_muvie(self):
open_file = filedialog.askopenfilename()
if open_file:
self._cam = VideoCapture(open_file)
self._prepare_for_word()
self._file_name['text'] = 'file name: ' + open_file.split('/')[-1]
class ProcessPipelineWithURL:
writeCurses = False
def __init__(self, cameraStreamURL, Pipeline):
self.url = cameraStreamURL
self.pipeline = Pipeline()
self.logger = logging.getLogger('pipeline-procesor')
self.stream = None
NetworkTables.setTeam(TEAM_NUMBER)
NetworkTables.initialize(server=NT_SERVER)
self.table = NetworkTables.getTable(NT_TABLE_NAME)
def readStreamFrame(self):
init = False
if not self.stream:
init = True
self.stream = VideoCapture(self.url)
connected, frame = self.stream.read()
if not connected:
self.logger.warning('Camera stream could not be read')
time.sleep(1)
self.stream.release()
self.stream = None
return None
else:
if init:
self.logger.debug('Stream status: %s', connected)
return frame
def sortTupleListByIdx(self, tupleList, idx):
return sorted(tupleList, key=lambda x: x[idx])
def initCurses(self):
self.writeCurses = True
self.scr = curses.initscr()
def processContour(self, contour):
self.logger.debug("Contour: \n %s", contour)
x_values = []
y_values = []
self.logger.debug("Getting x-values and y-values of contour")
for arr in contour:
x_values.append(arr[0][0])
y_values.append(arr[0][1])
self.logger.debug("Calculating max and min")
x_max = numpy.max(x_values)
y_max = numpy.max(y_values)
x_min = numpy.min(x_values)
y_min = numpy.min(y_values)
self.logger.debug("Calculating width and height")
width = x_max - x_min
height = y_max - y_min
self.logger.debug("Calculating center")
center = ((x_max + x_min) / 2, (y_max + y_min) / 2)
self.logger.debug("Contour height: %s, width: %s, center: %s", height,
width, center)
return width, height, center
def sendPipelineOutput(self):
self.logger.debug("start of sendPipelineOutput class")
idx = 0
self.logger.debug("get contour list from pipeline")
contour_list = getattr(self.pipeline, "filter_contours_output")
if len(contour_list) == 0:
return
if self.writeCurses:
self.scr.clear()
self.logger.debug("iterating through contour list")
for contour in contour_list:
n = "contour_%s" % idx
self.logger.debug("processing the contour")
width, height, center = self.processContour(contour)
self.logger.debug("send width, height, center to networktables")
self.table.putNumber(n + "_width", width)
self.table.putNumber(n + "_height", height)
self.table.putNumberArray(n + "_centerpoint", center)
self.logger.debug('Name: %s height: %s width: %s center: %s', n,
height, width, center)
idx += 1
if self.writeCurses:
for point in contour:
self.cursesTerminalWrite(point)
self.cursesTerminalWrite(center, char="X")
self.logger.debug('Centerpoint: (%s,%s)', center[1], center[0])
def cursesTerminalWrite(self, point, char="#"):
percent = tuple(map(operator.truediv, point, VIS_SIZE))
screenPercent = (percent[1], percent[0])
x, y = tuple(map(operator.mul, screenPercent, self.scr.getmaxyx()))
try:
self.scr.addstr(int(x), int(y), str(char))
self.scr.refresh()
except BaseException as er:
print(er)
def run(self):
self.logger.info('Attempting to process camera stream')
while True:
self.logger.debug("read frame")
frame = self.readStreamFrame()
if not frame is None:
self.logger.debug("Sending frame to GRIP pipeline")
self.pipeline.process(frame)
self.logger.debug("Start sending to rio")
self.sendPipelineOutput()
class GamePage(BasePage):
def __init__(self, window):
super(GamePage, self).__init__(window)
# ------------------------- 游戏是否结束 -------------------------#
self.hasGameOver = False
# ------------------------- 左右分数 -------------------------#
self.leftScore = 0
self.rightScore = 0
# ------------------------- 游戏模式 -------------------------#
# 检测人脸的个数
if BasePage.mode == MODE.SINGLE:
self.faceCount = 1
elif BasePage.mode == MODE.DOUBLE:
self.faceCount = 2
# ------------------------- 暂停 -------------------------#
# 是否暂停
self.isPause = False
# 暂停文字
# 标题文本
self.pauseSurface = self.middleFont.render('Pause', True, WHITE)
# 获取文本的矩形
self.pauseRect = self.pauseSurface.get_rect()
# 设置矩形中心点
self.pauseRect.center = (WINDOW_WIDTH / 2, WINDOW_HEIGHT / 2)
# ------------------------- 检测的人脸方框数据 -------------------------#
self.faceRects = []
# ------------------------- 存放球拍数据 -------------------------#
self.racketSurfaces = []
# ------------------------- 球拍 -------------------------#
# 乒乓球拍
self.racketSurface = image.load('image/pai.png')
# 球拍的宽度和高度
self.racketW = self.racketSurface.get_width()
self.racketH = self.racketSurface.get_height()
# ------------------------- 小球位置 -------------------------#
# 小球位置
self.pos_x = 300 # 矩形的X轴坐标
self.pos_y = 250 # 矩形的Y轴坐标
# 小球x和y轴分速度
self.vel_x = 24 # X轴分速度
self.vel_y = 12 # Y轴分速度
# ------------------------- 球 -------------------------#
# 乒乓球拍
ballSurface = image.load('image/ball.png')
# 球拍的宽度和高度
w = ballSurface.get_width()
h = ballSurface.get_height()
self.ballSurface = transform.scale(ballSurface, (w // 20, h // 20))
# 球矩形
self.ballRect = self.ballSurface.get_rect()
# 球的宽度和高度
self.ballW = self.ballRect.width
self.ballH = self.ballRect.height
# 设置矩形中心点
self.ballRect.topleft = (self.pos_x, self.pos_y)
# tensorflow检测类
self.tfDetector = TensoflowFaceDector(PATH_TO_CKPT)
# ---------------------- 读取配置获取摄像头id ---------------------- #
f = open('config/config.txt')
lines = f.readlines()
for line in lines:
if line.startswith('camera_index'):
CAMERA_INDEX = int(line[line.index(':') + 1:])
elif line.startswith('game_over_score'):
self.game_over_score = int(line[line.index(':') + 1:])
print('---------------摄像头id:', CAMERA_INDEX)
# 摄像头
self.camera = VideoCapture(CAMERA_INDEX)
self.camera.set(3, WINDOW_WIDTH)
self.camera.set(4, WINDOW_HEIGHT)
def getCamFrame(self):
'''
摄像头获取一帧数据
:param camera: 摄像头
:return:
'''
# 清空人脸方框数据
self.faceRects.clear()
if not self.camera.isOpened():
print('摄像头没有打开')
return None
# 获取图像
retval, frame = self.camera.read()
if retval:
print('成功-------------------------------------------------------')
frame = cvtColor(frame, COLOR_BGR2RGB)
[h, w] = frame.shape[:2]
# tensorflow检测人脸
(boxes, scores, classes,
num_detections) = self.tfDetector.run(frame)
boxes = squeeze(boxes)
scores = squeeze(scores)
for i in range(0, self.faceCount):
if scores[i] > 0.3:
ymin, xmax, ymax, xmin = boxes[i, 0], boxes[i, 1], boxes[
i, 2], boxes[i, 3]
left, right, top, bottom = (xmin * w, xmax * w, ymin * h,
ymax * h)
# 处理左右坐标
left = WINDOW_WIDTH - left
right = WINDOW_WIDTH - right
self.faceRects.append((left, right, top, bottom))
print('--------left', left, ' right', right)
# cv2.rectangle(frame, (int(left), int(top)), (int(right), int(bottom)), (0, 255, 255), 2)
frame = rot90(frame)
frame = surfarray.make_surface(frame)
return frame
else:
print('失败-------------------------------------------------------')
return None
def handleSingleModeCollision(self):
'''单人模式碰撞'''
for _, rackRect in self.racketSurfaces:
left = rackRect[0]
top = rackRect[1]
width = rackRect[2]
height = rackRect[3]
# 右侧矩形
rightRect = Rect(left + width - 1, top, 1, height * 2 // 3)
# 小球从右侧过来 是否和右侧碰撞
if rightRect.colliderect(self.ballRect) and self.vel_x < 0:
self.vel_x = abs(self.vel_x)
self.rightScore += 1
return
# 左侧矩形
leftRect = Rect(left, top, 1, height * 2 // 3)
# 从左侧过来 是否和左侧碰撞
if leftRect.colliderect(self.ballRect) and self.vel_x > 0:
self.vel_x = -abs(self.vel_x)
self.leftScore += 1
return
def handleDoubleModeCollision(self):
'''双人模式碰撞'''
for _, rackRect in self.racketSurfaces:
left = rackRect[0]
top = rackRect[1]
width = rackRect[2]
height = rackRect[3]
# 中间的x
centerX = left + width / 2
# 判断左右位置
if centerX < WINDOW_WIDTH / 2:
# print('左侧--------------------------------------------------------------------','left',left,' right',right)
# 左侧
# 右侧矩形
rightRect = Rect(left + width - 1, top, 1, height * 2 // 3)
# 小球从右侧过来 是否和右侧碰撞
if rightRect.colliderect(self.ballRect) and self.vel_x < 0:
self.vel_x = abs(self.vel_x)
return
else:
# print('右侧--------------------------------------------------------------------')
# 右侧
# 左侧矩形
leftRect = Rect(left, top, 1, height * 2 // 3)
# 从左侧过来 是否和左侧碰撞
if leftRect.colliderect(self.ballRect) and self.vel_x > 0:
self.vel_x = -abs(self.vel_x)
return
def handleCollision(self):
'''处理球拍和小球碰撞'''
if self.mode == MODE.SINGLE:
self.handleSingleModeCollision()
elif self.mode == MODE.DOUBLE:
self.handleDoubleModeCollision()
def display(self):
# 是否暂停
if self.isPause:
self.window.blit(self.pauseSurface, self.pauseRect)
return
# 获取一帧画面
frame = self.getCamFrame()
# 如果没有画面不做处理
if not frame:
print('没有获取画面')
return
# 展示画面
self.window.blit(frame, (0, 0))
# 展示人脸球拍数据
self.showRacket()
# 处理球拍和小球碰撞
self.handleCollision()
# 展示乒乓球
self.showBall()
# 显示分数
self.showScore()
if not self.hasGameOver and self.mode == MODE.DOUBLE:
# 游戏结束判断
self.gameOver()
def gameOver(self):
'''游戏结束判断'''
if self.leftScore >= self.game_over_score or self.rightScore >= self.game_over_score:
# 游戏结束标示
self.hasGameOver = True
# 再展示一次
self.display()
# 刷新
display.flip()
# 修改页面标示
BasePage.pageType = PAGETYPE.GAMEOVER
def showSingleModeScore(self):
'''单人模式分数'''
# 标题文本
self.leftScoreSurface = self.middleFont.render(
'左击:{}'.format(self.leftScore), True, WHITE)
self.rightScoreSurface = self.middleFont.render(
'右击:{}'.format(self.rightScore), True, WHITE)
# 显示分数
self.window.blit(self.leftScoreSurface, (100, 50))
self.window.blit(self.rightScoreSurface, (WINDOW_WIDTH - 200, 50))
def showDoubleModeScore(self):
'''双人模式分数'''
# 标题文本
self.leftScoreSurface = self.middleFont.render(
'{}'.format(self.leftScore), True, WHITE)
self.rightScoreSurface = self.middleFont.render(
'{}'.format(self.rightScore), True, WHITE)
# 显示分数
self.window.blit(self.leftScoreSurface, (100, 50))
self.window.blit(self.rightScoreSurface, (WINDOW_WIDTH - 100, 50))
def showScore(self):
'''显示分数'''
# 单人模式不显示分数
if self.mode == MODE.SINGLE:
self.showSingleModeScore()
if self.mode == MODE.DOUBLE:
self.showDoubleModeScore()
def constructRacketDatas(self):
'''根据人脸方框构建球拍数据'''
# 清空数据
self.racketSurfaces.clear()
for face in self.faceRects:
left = face[0]
right = face[1]
top = face[2]
bottom = face[3]
w = right - left
h = bottom - top
# 中心点
centerP = left + w // 2, top + h // 2
# 左侧赢 右侧不用展示球拍 右侧赢 左侧不需要展示
if (self.leftScore >= 5 and centerP[0] > WINDOW_WIDTH / 2) or (
self.rightScore >= 5 and centerP[0] < WINDOW_WIDTH / 2):
continue
# 缩小系数
if w > h:
# 缩小倍数
rf = self.racketH / h
else:
rf = self.racketW / w
print('缩放系数', rf)
racketSurface = transform.scale(
self.racketSurface,
(int(self.racketW / rf), int(self.racketH / rf)))
# 球拍举行
racketRect = racketSurface.get_rect()
# 设置矩形中心点
racketRect.center = centerP
# self.window.blit(racketSurface, racketRect)
self.racketSurfaces.append((racketSurface, racketRect))
def showRacket(self):
'''显示球拍'''
# 构建球拍数据
self.constructRacketDatas()
# 显示到窗口上
for ele in self.racketSurfaces:
self.window.blit(ele[0], ele[1])
def showBall(self):
'''显示乒乓球'''
# 移动矩形
self.pos_x += self.vel_x
self.pos_y += self.vel_y
# 保证矩形能待在屏幕内
if self.pos_x > WINDOW_WIDTH - self.ballW:
self.vel_x = -self.vel_x
if self.mode == MODE.DOUBLE:
# 左边得分
self.leftScore += 1
elif self.pos_x < 0:
self.vel_x = -self.vel_x
if self.mode == MODE.DOUBLE:
# 右边得分
self.rightScore += 1
if self.pos_y > WINDOW_HEIGHT - self.ballH or self.pos_y < 0:
self.vel_y = -self.vel_y
# 修改小球的x和y坐标
self.ballRect.topleft = (self.pos_x, self.pos_y)
# 画乒乓球
self.window.blit(self.ballSurface, self.ballRect)
def keyDown(self, key):
'''
按下的事件
:param key:按下的键
:return:
'''
if key == K_SPACE:
# 空格键 暂停
self.isPause = not self.isPause
elif key == K_TAB:
BasePage.pageType = PAGETYPE.SPLASH