def main(argv):
global isNewPoint
global newPoint
img = cv2.imread("Penguins.jpg")
img = u.resize(img)
winName="watershed Testes"
imgPrev = np.copy(img)
cv2.namedWindow(winName)
cv2.imshow(winName, imgPrev)
cv2.setMouseCallback(winName, onMouse)
op='f'
fgRect=[]
bgRect=[]
while op!='p' and op!='P':
op = cv2.waitKey(50)
if isNewPoint == 1:
isNewPoint=0
if op == 'f' or op == 'F':
fgRect+=[newPoint]
elif op == 'f' or op == '':
pass
cv2.destroyAllWindows()
def update(self, dt):
ret, frame = self.capture.read()
if frame is None:
sys.exit()
frame = utilities.resize(frame, 480, 480)
# convert it to texture
buf1 = cv2.flip(frame, 0)
buf = buf1.tostring()
image_texture = Texture.create(
size=(frame.shape[1], frame.shape[0]), colorfmt='bgr')
image_texture.blit_buffer(buf, colorfmt='bgr', bufferfmt='ubyte')
# display image from the texture
self.texture = image_texture
def test_resize(self): image = decode_base64_image(TEST_BINARY_DATA) img = images.Image(resize(image)) self.assertEqual(img.width, 200) self.assertEqual(int(img.height), 170)
import imutils
from queue import Queue
firstFrame = None
consecFrames = 0
kcw = KeyClipWriter(bufSize=128)
print("[INFO] Camera is warming up... Please Wait...")
# Use threads to read from camera
cap = VideoStream(src=0).start()
time.sleep(3)
fourcc = cv2.VideoWriter_fourcc(*'MJPG')
while True:
frame = cap.read()
frame = utilities.resize(frame, width=500)
updateConsecFrame = True
frame = cv2.flip(frame, 1)
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (21, 21), 0)
if firstFrame is None:
firstFrame = gray
cv2.imshow("First Frame", firstFrame)
continue
cv2.imshow("Original Video", frame)
difference = cv2.absdiff(gray, firstFrame)
thresh = cv2.threshold(difference, 20, 255, cv2.THRESH_BINARY)[1]
thresh = cv2.dilate(thresh, None, iterations=2)
contours = cv2.findContours(thresh, cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(contours)
def frames():
# load our serialized model from disk
print("[INFO] loading model...")
net = cv2.dnn.readNetFromCaffe(PTOTO_PATH, CAFFE_PATH)
camera = cv2.VideoCapture(Camera.video_source, cv2.CAP_DSHOW)
kcw = KeyClipWriter(bufSize=128)
fourcc = cv2.VideoWriter_fourcc(*'MJPG')
consecFrames = 0
md = MotionDetector(accumWeight=0.4)
COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))
total = 0
frameCount = 16
frame_rate = 16
# previous_time = 0
if not camera.isOpened():
raise RuntimeError('Could not start camera.')
while True:
# print("Name of video file from original function: ", Camera.name_of_videoFile)
updateConsecFrame = True
# read current frame
# time_elapsed = time() - previous_time
_, frame = camera.read()
# if time_elapsed > 1./frame_rate:
# previous_time = time()
# fps = FPS().start()
(h, w) = frame.shape[:2]
blob = cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)), 0.007843, (300, 300), 127.5)
frame = cv2.flip(frame, 1)
frame = utilities.resize(frame,width=350)
frame_for_video = frame.copy()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (7, 7), 0)
if total > frameCount:
motion = md.detect(gray)
if motion is not None:
(thresh, (minX, minY, maxX, maxY)) = motion
cv2.rectangle(frame, (minX, minY), (maxX, maxY), (0, 0, 255), 2)
net.setInput(blob)
detections = net.forward()
consecFrames = 0
for i in np.arange(0, detections.shape[2]):
confidence = detections[0, 0, i, 2]
if confidence > MIN_CONFIDENCE:
idx = int(detections[0, 0, i, 1])
if CLASSES[idx] in IGNORE:
continue
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
(startX, startY, endX, endY) = box.astype("int")
label = "{}: {:.2f}%".format(CLASSES[idx], confidence * 100)
cv2.rectangle(frame, (startX, startY), (endX, endY),
COLORS[idx], 2)
y = startY - 15 if startY - 15 > 15 else startY + 15
cv2.putText(frame, label, (startX, y), cv2.FONT_HERSHEY_SIMPLEX, 0.5, COLORS[idx], 2)
if not kcw.recording:
timestamp = datetime.datetime.now()
name_of_videoFile = "{}-{}.avi".format("output",timestamp.strftime("%Y-%m-%d-%H-%M-%S"))
kcw.start(name_of_videoFile, fourcc, 30)
if updateConsecFrame:
consecFrames += 1
# # update the key frame clip buffer
kcw.update(frame_for_video)
# if we are recording and reached a threshold on consecutive
# number of frames with no action, stop recording the clip
if kcw.recording and consecFrames == 64:
print("Name from Original Function: ", name_of_videoFile)
subprocess.run(['python', 'print_test.py', name_of_videoFile])
kcw.finish()
# encode as a jpeg image and return it
md.update(gray)
total += 1
cv2.waitKey(10)
# fps.update()
yield cv2.imencode('.jpg', frame)[1].tobytes()
def getdata():
trainsize = 5000
testsize = 1000
folder = ""
# read training labels
with open(
os.path.dirname(os.path.dirname(os.path.abspath(__file__))) +
"\dataset\\" + "train-labels.idx1-ubyte", "rb") as f:
data = f.read(4) # magic number
data = f.read(4) # number of samples
training_samples = struct.unpack('>HH', data)[1]
traininglabels = np.empty(trainsize)
for i in range(0, trainsize):
data = f.read(1)
traininglabels[i] = int(struct.unpack('>B', data)[0])
# read training images
with open(
os.path.dirname(os.path.dirname(os.path.abspath(__file__))) +
"\dataset\\" + "train-images.idx3-ubyte", "rb") as f:
data = f.read(4) # magic number
data = f.read(4) # number of samples
trainingsample = np.zeros(784)
trainingsamples = np.empty((trainsize, 196))
data = f.read(4) # y dimension
data = f.read(4) # x dimension
for i in range(0, trainsize):
for j in range(0, 784):
data = f.read(1)
trainingsample[j] = util.round(struct.unpack('>B', data)[0])
trainingsamples[i] = util.subsample(util.resize(trainingsample),
14)
# read test labels
with open(
os.path.dirname(os.path.dirname(os.path.abspath(__file__))) +
"\dataset\\" + "t10k-labels.idx1-ubyte", "rb") as f:
data = f.read(4) # magic number
data = f.read(4) # number of samples
test_samples = struct.unpack('>HH', data)[1]
testlabels = np.empty(testsize)
for i in range(0, testsize):
data = f.read(1)
testlabels[i] = int(struct.unpack('>B', data)[0])
# read test images
with open(
os.path.dirname(os.path.dirname(os.path.abspath(__file__))) +
"\dataset\\" + "t10k-images.idx3-ubyte", "rb") as f:
data = f.read(4) # magic number
data = f.read(4) # number of samples
testsample = np.zeros(784)
testsamples = np.empty((testsize, 196))
data = f.read(4) # y dimension
data = f.read(4) # x dimension
for i in range(0, testsize):
for j in range(0, 784):
data = f.read(1)
testsample[j] = util.round(struct.unpack('>B', data)[0])
testsamples[i] = util.subsample(util.resize(testsample), 14)
y_train = traininglabels
X_train = trainingsamples
y_test = testlabels
X_test = testsamples
return X_train, X_test, y_train, y_test
def update(self, dt):
ret, frame = self.capture.read()
# if frame is None:
# print("[Info] Frame not Read... Leaving....")
# sys.exit()
if ret:
frame = utilities.resize(frame, 480, 480)
# convert it to texture
(h, w) = frame.shape[:2]
blob = cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)), 1.0,
(300, 300), (104.0, 177.0, 123.0))
# pass the blob through the network and obtain the detections and
# predictions
net.setInput(blob)
detections = net.forward()
# loop over the detections
for i in range(0, detections.shape[2]):
# extract the confidence (i.e., probability) associated with the
# prediction
confidence = detections[0, 0, i, 2]
# filter out weak detections
if confidence > 0.85:
# compute the (x, y)-coordinates of the bounding box for
# the face and extract the face ROI
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
(startX, startY, endX, endY) = box.astype("int")
# ensure the detected bounding box does fall outside the
# dimensions of the frame
startX = max(0, startX)
startY = max(0, startY)
endX = min(w, endX)
endY = min(h, endY)
# extract the face ROI and then preproces it in the exact
# same manner as our training data
face = frame[startY:endY, startX:endX]
if len(face) == 0:
continue
face = cv2.resize(face, (32, 32))
face = face.astype("float") / 255.0
face = img_to_array(face)
face = np.expand_dims(face, axis=0)
# pass the face ROI through the trained liveness detector
# model to determine if the face is "real" or "fake"
preds = model.predict(face)[0]
j = np.argmax(preds)
label = le.classes_[j]
if label == "Real":
rect_color = (255, 128, 255)
"""
# Important Note
# Recognition Runs here
# Do not forgot to import face recognition
username = RecognizeMe(frame)
if username != "unknown":
# following lines should be shifted to inside this if block
# Then comment/ remove line username = '******'
"""
ev = OnFaceRecognition()
ev.bind(on_recognize=custom_event_callback)
username = '******'
ev.goHome('Real', username)
self.capture.release()
else:
rect_color = (0, 0, 255)
# draw the label and bounding box on the frame
label = "{}: {:.2f}".format(label, preds[j])
cv2.rectangle(frame, (startX - 1 , startY - 20), (startX - int((startX - endX) / 1.5), startY),
rect_color, -1)
cv2.putText(frame, label, (startX, startY - 5),
cv2.FONT_HERSHEY_PLAIN, 0.80, (255, 255, 255), 1, cv2.LINE_AA)
cv2.rectangle(frame, (startX, startY), (endX, endY),
rect_color, 2)
buf1 = frame.copy()
buf1 = cv2.flip(buf1, 0)
buf = buf1.tostring()
image_texture = Texture.create(
size=(frame.shape[1], frame.shape[0]), colorfmt='bgr')
image_texture.blit_buffer(buf, colorfmt='bgr', bufferfmt='ubyte')
# display image from the texture
self.texture = image_texture
def frames():
camera = cv2.VideoCapture(Camera.video_source)
if not camera.isOpened():
raise RuntimeError('Could not start camera.')
while True:
# read current frame
_, frame = camera.read()
if frame is None:
from app import app
with app.app_context():
# return redirect(url_for('ctrl_video'))
# from flask import redirect, url_for
# redirect = url_for('ctrl_video')
msg = {"msg": "success"}
r = requests.post("http://127.0.0.1:5000/", json=msg)
break
# grab the frame dimensions and convert it to a blob
frame = utilities.resize(frame, 400)
(h, w) = frame.shape[:2]
blob = cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)), 1.0,
(300, 300), (104.0, 177.0, 123.0))
# pass the blob through the network and obtain the detections and
# predictions
net.setInput(blob)
detections = net.forward()
# loop over the detections
for i in range(0, detections.shape[2]):
# extract the confidence (i.e., probability) associated with the
# prediction
confidence = detections[0, 0, i, 2]
# filter out weak detections
if confidence > 0.85:
# compute the (x, y)-coordinates of the bounding box for
# the face and extract the face ROI
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
(startX, startY, endX, endY) = box.astype("int")
# ensure the detected bounding box does fall outside the
# dimensions of the frame
startX = max(0, startX)
startY = max(0, startY)
endX = min(w, endX)
endY = min(h, endY)
# extract the face ROI and then preproces it in the exact
# same manner as our training data
face = frame[startY:endY, startX:endX]
if len(face) == 0:
continue
face = cv2.resize(face, (32, 32))
face = face.astype("float") / 255.0
face = img_to_array(face)
face = np.expand_dims(face, axis=0)
# pass the face ROI through the trained liveness detector
# model to determine if the face is "real" or "fake"
preds = model.predict(face)[0]
j = np.argmax(preds)
label = le.classes_[j]
if label == "Real":
rect_color = (255, 128, 255)
else:
rect_color = (0, 0, 255)
# draw the label and bounding box on the frame
label = "{}: {:.2f}".format(label, preds[j])
cv2.rectangle(frame, (startX - 1 , startY - 20), (startX - int((startX - endX) / 1.5), startY),
rect_color, -1)
cv2.putText(frame, label, (startX, startY - 5),
cv2.FONT_HERSHEY_PLAIN, 0.80, (255, 255, 255), 1, cv2.LINE_AA)
cv2.rectangle(frame, (startX, startY), (endX, endY),
rect_color, 2)
cv2.waitKey(1)
# encode as a jpeg image and return it
yield cv2.imencode('.jpg', frame)[1].tobytes()
# Ignoring Unwanted classes
COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))
# load our serialized model from disk
print("[INFO] loading model...")
net = cv2.dnn.readNetFromCaffe(PTOTO_PATH, CAFFE_PATH)
# load the input image and construct an input blob for the image
# by resizing to a fixed 300x300 pixels and then normalizing it
# (note: normalization is done via the authors of the MobileNet SSD
# implementation)
cap = cv2.VideoCapture(0, cv2.CAP_DSHOW)
while True:
ret, image = cap.read()
image = utilities.resize(image, 350, 350)
(h, w) = image.shape[:2]
blob = cv2.dnn.blobFromImage(cv2.resize(image, (300, 300)), 0.007843,
(300, 300), 127.5)
# pass the blob through the network and obtain the detections and
# predictions
print("[INFO] computing object detections...")
net.setInput(blob)
detections = net.forward()
# loop over the detections
for i in np.arange(0, detections.shape[2]):
# extract the confidence (i.e., probability) associated with the
# prediction
confidence = detections[0, 0, i, 2]
# filter out weak detections by ensuring the `confidence` is