frame = video_getter.frame.copy()
frame = imutils.resize(frame, width=500)
rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
(H, W) = frame.shape[:2]
rects = []
if cpt % skipped_frames == 0:
recon = []
fotos = []
ps = []
trackers = []
devices = []
detections = get_faces(detector, embedder, sp,frame, D_PROB, fa)
#[(face,vector,coordenada,imagen_completa)]
face_data = [(*face, *recognize(face[1], recognizer, le, R_PROB)) for face in detections]
#[(face,vector,coordenada,imagen_completa, nombre, prob)]
for item in face_data:
#Listas con nombres de reconocidos
recon.append(item[4])
fotos.append(item[0])
ps.append(item[5])
[devices.append(1) for i in range(len(face_data))]
for face in detections:
(startX, startY, endX, endY) = face[2]
tracker = dlib.correlation_tracker()
rect = dlib.rectangle(startX, startY, endX, endY)
tracker.start_track(rgb, rect)
trackers.append(tracker)
# loop over the trackers
for item in face_data:
rects_esp = []
#print(cpt)
if cpt % skipped_frames== 0:
recon_pi = []
fotos_pi = []
ps_pi = []
trackers_pi = []
trackers_esp32 = []
recon_esp32 = []
fotos_esp32 = []
ps_esp32 = []
detections = get_faces(detector, embedder, frame, 0.5, fa)
#[(face,vector,coordenada,imagen_completa)]
face_data = [(*face, *recognize(face[1], recognizer, le, 0.65)) for face in detections]
#[(face,vector,coordenada,imagen_completa, nombre, prob)]
for item in face_data:
#Listas con nombres de reconocidos
recon_pi.append(item[4])
fotos_pi.append(item[0])
ps_pi.append(item[5])
#[devices.append(1) for i in range(len(face_data))]
for face in detections:
(startX, startY, endX, endY) = face[2]
tracker = dlib.correlation_tracker()
rect = dlib.rectangle(startX, startY, endX, endY)
tracker.start_track(rgb, rect)
trackers_pi.append(tracker)
# loop over the trackers
frame_esp32 = esp32_frame('grupo14.duckdns.org', 1228)
def main_core(args, frame_queue, pframe_queue):
# load our serialized face detector from disk
print("[INFO] loading face detector...")
protoPath = os.path.sep.join([args["detector"], "deploy.prototxt"])
modelPath = os.path.sep.join([args["detector"],
"res10_300x300_ssd_iter_140000.caffemodel"])
detector = cv2.dnn.readNetFromCaffe(protoPath, modelPath)
predictor = dlib.shape_predictor(args["shape_predictor"])
fa = FaceAligner(predictor, desiredFaceWidth=256)
# load our serialized face embedding model from disk
print("[INFO] loading face recognizer...")
embedder = cv2.dnn.readNetFromTorch(args["embedding_model"])
# load the face embeddings
print("[INFO] loading face embeddings...")
data = pickle.loads(open(args["embeddings"], "rb").read())
recognizer, le = train(data)
time.sleep(1.0)
# initialize the video stream, then allow the camera sensor to warm up
## Set Threding to start filming
video_getter = VideoGet(frame_queue=frame_queue, src=0, name='Video Getter')
time.sleep(1.0)
# print('[INFO] Starting VideoGet...')
video_getter.start()
time.sleep(1.0)
ct = CentroidTracker(maxDisappeared=25, maxDistance=75)
trackers = []
trackableObjects = {}
out = 0
skipped_frames = 2
out_prev = 0
cpt=0;
exitbool = False
fps_count = FPS().start()
while True:
frame = video_getter.frame.copy()
# frame = esp32_frame("grupo14.duckdns.org", 1228)
# if frame is None:
# continue
# frame = np.array(frame)
frame = imutils.resize(frame, width=500)
rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
(H, W) = frame.shape[:2]
rects = []
#print(cpt)
if cpt % skipped_frames== 0:
recon = []
fotos = []
ps = []
trackers = []
detections = get_faces(detector, embedder, frame, 0.5, fa)
#[(face,vector,coordenada,imagen_completa)]
face_data = [(*face, *recognize(face[1], recognizer, le, 0.65)) for face in detections]
#[(face,vector,coordenada,imagen_completa, nombre, prob)]
for item in face_data:
#Listas con nombres de reconocidos
recon.append(item[4])
fotos.append(item[0])
ps.append(item[5])
for face in detections:
(startX, startY, endX, endY) = face[2]
tracker = dlib.correlation_tracker()
rect = dlib.rectangle(startX, startY, endX, endY)
tracker.start_track(rgb, rect)
# add the tracker to our list of trackers so we can
# utilize it during skip frames
trackers.append(tracker)
# loop over the trackers
else:
for tracker in trackers:
# update the tracker and grab the updated position
tracker.update(rgb)
pos = tracker.get_position()
# unpack the position object
startX = int(pos.left())
startY = int(pos.top())
endX = int(pos.right())
endY = int(pos.bottom())
# add the bounding box coordinates to the rectangles list
rects.append((startX, startY, endX, endY))
objects, names, images, probabilities = ct.update(rects,recon, fotos, ps)
# loop over the tracked objects
for (objectID, centroid),(ID, name),(I,im),(D,prob) in zip(objects.items(),
names.items(), images.items(), probabilities.items()):
to = trackableObjects.get(objectID, None)
if to is None:
to = TrackableObject(objectID, centroid, name, im, prob)
else:
y = [c[1] for c in to.centroids]
direction = centroid[1] - np.mean(y)
to.centroids.append(centroid)
# Si es que salio
# check to see if the object has been counted or not
if not to.counted and direction > 0 and centroid[1] > H - 50:
to.out = True
to.counted = True
# store the trackable object in our dictionary
trackableObjects[objectID] = to
#Coordinamos el paquete de envio
## Envio de mails
if not to.sent:
paquete = [to.prob, to.pic, to.reconocido, to.out]
if to.reconocido:
#enviar mail
paquete.append(to.name)
else:
paquete.append('unknown {}'.format(objectID))
# enviar mails
##Paquete a enviar
to.sent = True
for item in face_data:
print('Reconocido ',item[4])
# frame = draw_frame(frame, item)
fps_count.update()
cpt += 1
out_prev = out
# if cpt > 250:
# video_getter.stop()
# break
exitbool = 0# show_frame(frame)
if exitbool or cpt > 50:
# SV.stop()
fps_count.stop()
print("[INFO] elasped time fps processed: {:.2f}".format(fps_count.elapsed()))
print("[INFO] approx. processed FPS: {:.2f}".format(fps_count.fps()))
time.sleep(1)
video_getter.stop()
# db_client.close()
break
rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
(H, W) = frame.shape[:2]
rects = []
#print(cpt)
if cpt % skipped_frames == 0:
recon = []
fotos = []
ps = []
trackers = []
trackers_esp32 = []
devices = []
detections = get_faces(detector, embedder, frame, 0.5, fa)
#[(face,vector,coordenada,imagen_completa)]
face_data = [(*face, *recognize(face[1], recognizer, le, 0.65))
for face in detections]
#[(face,vector,coordenada,imagen_completa, nombre, prob)]
for item in face_data:
#Listas con nombres de reconocidos
recon.append(item[4])
fotos.append(item[0])
ps.append(item[5])
[devices.append(1) for i in range(len(face_data))]
for face in detections:
(startX, startY, endX, endY) = face[2]
tracker = dlib.correlation_tracker()
rect = dlib.rectangle(startX, startY, endX, endY)
tracker.start_track(rgb, rect)
trackers.append(tracker)
# loop over the trackers
for name in imagePaths:
Coincidencias = 0
Caras = 0
print('Persona a evaluar : {}'.format(name))
print('Prepare imagen')
input('Presione cualquier boton')
for j in range(4):
for i in range(frames_per_person):
frame = video_getter.frame.copy()
#frame = imutils.resize(frame, width=500)
detections = get_faces(detector, embedder, sp, frame,
confianza_dec, fa)
#[(face,vector,coordenada,imagen_completa)]
face_data = [(*face,
*recognize(face[1], recognizer, le, confianza_recon))
for face in detections]
#[(face,vector,coordenada,imagen_completa, nombre, prob)]
for item in face_data:
print(item[4])
if item[4] == name:
Coincidencias += 1
Caras += 1
show_frame(frame)
fps_count.update()
input('Cambio de foto')
calculos[name] = (Coincidencias, Caras)
print(calculos)
fps_count.stop()
print('Persona a evaluar : {}'.format(name))
while True:
(grabbed, frame) = stream.read()
# if the frame was not grabbed, then we have reached the
# end of the stream
if not grabbed:
stream.release()
time.sleep(2.0)
break
frame = imutils.resize(frame, width=450)
detections = get_faces(detector, embedder, sp, frame, confianza_dec,
fa)
#[(face,vector,coordenada,imagen_completa)]
face_data = [(*face, *recognize(face[1], model, le, confianza_recon))
for face in detections]
#[(face,vector,coordenada,imagen_completa, nombre, prob)]
for item in face_data:
x_test.append(item[1])
y_test.append(item[4])
y_train.append(name)
prob.append(item[5])
if item[4] == name:
Coincidencias += 1
elif item[4] != 'unknown':
Falsas_coincidencias += 1
Caras += 1
fps_count.update()
calculos[name] = (Coincidencias, Caras, Falsas_coincidencias,
np.mean(prob))