OpenCV_Recognizer/detect_faces.py at dev · AlexRioja/OpenCV_Recognizer · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
import cv2
from time import sleep
import pickle

wait_4_camera=True

face_classifier = cv2.CascadeClassifier('resources/classifiers/haarcascade_frontalface_alt2.xml')
eye_classifier = cv2.CascadeClassifier('resources/classifiers/haarcascade_eye.xml')


fisher_recognizer = cv2.face.FisherFaceRecognizer_create()
lbphf_recognizer = cv2.face.LBPHFaceRecognizer_create()
eigen_recognizer= cv2.face.EigenFaceRecognizer_create()

fisher_recognizer.read("recognizer/trainer_fisher.yml")
lbphf_recognizer.read("recognizer/trainer_lbphf.yml")
eigen_recognizer.read("recognizer/trainer_eigen.yml")

labels={}

def image_preprocessing(img):
	#ampliamos el espectro de la imagen para normalizar la intesidad de pixel (mayor contraste, mejor extracción de features)
	return cv2.equalizeHist(img)

with open("recognizer/labels.pickle", "rb") as f:
	inv_labels=pickle.load(f)
	labels={v:k for k,v in inv_labels.items()}#invertimos las claves y los valores de posicion


while wait_4_camera:
	video_interface = cv2.VideoCapture(0)
	#Si no tenemos camarita...
	if not video_interface.isOpened():
		print('No se detecta cámara, probando de nuevo en 5 segundos!')
		sleep(5)
	else:
		print('Iniciando sistema de reconocimiento')
		wait_4_camera=False

#___________________A partir de aquí se realiza el reconocimiento___________________

while True:
	# Leemos el video frame a frame
	ret, frame = video_interface.read()
	#Lo pasamos a escala de grises
	gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
	#Detectamos las caras
	faces = face_classifier.detectMultiScale(
		gray,
		scaleFactor=1.1,
		minNeighbors=5,
		minSize=(50, 50)
	)

	# Dibujamos el rectangulito en las caras
	for (x, y, w, h) in faces:
		frame=cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 150), 2)
		#RegionOfInterest--->la carita buena
		roi_gray = gray[y:y+h, x:x+w]
		roi_color = frame[y:y+h, x:x+w]
		processed_img=image_preprocessing(roi_gray)
		if processed_img.shape < (50,50):
			processed_img=cv2.resize(processed_img, (50,50),interpolation=cv2.INTER_AREA)
		else:
			processed_img=cv2.resize(processed_img, (50,50),interpolation=cv2.INTER_CUBIC)
		#-------------FISHER
		id_, confidence= fisher_recognizer.predict(processed_img)

		font = cv2.FONT_ITALIC
		cv2.putText(frame, "Fisher->"+labels[id_],(x,y-5),font, 1, (255,255,255),2,cv2.LINE_AA)
		#-------------LBPHF
		id_, confidence= lbphf_recognizer.predict(processed_img)
		#if confidence<100:
		cv2.putText(frame, "LBPHF->"+labels[id_]+"--Confidence: "+str(round(155-confidence))+" %",(x,y-35),font, 1, (255,255,255),2,cv2.LINE_AA)
		#else:
		#	cv2.putText(frame, "LBPHF->Unknown",(x,y-35),font, 1, (255,255,255),2,cv2.LINE_AA)
		#-------------EIGEN
		id_, confidence= lbphf_recognizer.predict(processed_img)
		cv2.putText(frame, "Eigen->"+labels[id_],(x,y-65),font, 1, (255,255,255),2,cv2.LINE_AA)
		eyes = eye_classifier.detectMultiScale(roi_gray, minNeighbors=8, minSize=(50,50))
		for (ex,ey,ew,eh) in eyes:
			cv2.rectangle(roi_color,(ex,ey),(ex+ew,ey+eh),(255,0,0),2)

	# Sacamos una ventana con el video
	cv2.putText(frame, "Q para salir",(5,frame.shape[0]-5),cv2.FONT_HERSHEY_PLAIN, 1.3, (66,53,243), 2, cv2.LINE_AA)
	cv2.imshow('Video', frame)
	#Rompemos si pretamos 'q'
	if cv2.waitKey(1) & 0xFF == ord('q'):
		break

# Liberamos la interfaz de video y destruimos las ventanas creadas
print('Saliendo...')
video_interface.release()
cv2.destroyAllWindows()