def face():
speak('Processing Facial Recognition')
test_img = cv2.imread('saved/{}.jpg'.format(always))
faces_detected, gray_img = fr.faceDetection(test_img)
speak("I found your data from our system")
faces, faceID = fr.labels_for_training_data('test')
face_recognizer = fr.train_classifier(faces, faceID)
face_recognizer.write('trainingData.yml')
users_list()
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) #predicting the label of given image
print("Confidence:", confidence)
print("label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
if (
confidence > 37
): #If confidence more than 37 then don't print predicted face text on screen
welcome()
break
fr.put_text(test_img, predicted_name, x, y)
resized_img = cv2.resize(test_img, (1000, 1000))
cv2.imshow("Face", resized_img)
cv2.waitKey(0) #Waits indefinitely until a key is pressed
cv2.destroyAllWindows
def checkIn(name):
global db
print("About to take a Picture")
path = "/home/pi/Downloads/" + name + ".jpg"
with picamera.PiCamera() as camera:
camera.resolution = (1280, 720)
camera.capture(path)
print("Picture taken.")
test_img = cv2.imread("/home/pi/Desktop/Face Recognition/TestImages/" +
name + ".jpg")
faces_detected, gray_img = fr.faceDetection(test_img)
print("faces detected: ", faces_detected)
#use for testing purpose:-
face_recognizer = cv2.createLBPHFaceRecognizer()
face_recognizer.load('/home/pi/Desktop/Project3/trainingData.yml')
predicted_name = ""
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(roi_gray)
print("confidence: ", confidence)
print("label: ", label)
fr.draw_rect(test_img, face)
predicted_name = namesAndPath[label]
fr.put_text(test_img, predicted_name, x, y)
print(predicted_name + " You are checked-in successfully")
db[predicted_name] = datetime.datetime.utcnow()
return
def labeltrans(x1, x2):
x1 = int(x1)
ctr = 0
nctr = 0
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read('trainingData.yml')
name = {
0: "Aakarsh Anubhav",
1: "Soham Samanta",
2: "Kondala Snehasis Rao",
3: "Apurva Shahabadi",
4: "Shivam Singh",
5: "Kunwar Pratyush",
6: "Harsh Anand",
7: "Ch. Harika",
8: "Maasid Zafar"
}
cap = cv2.VideoCapture(0)
while ctr <= 21:
ret, test_img = cap.read()
faces_detected, gray_img = fr.faceDetection(test_img)
for (x, y, w, h) in faces_detected:
cv2.rectangle(test_img, (x, y), (x + w, y + h), (0, 255, 0), 2)
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + w, x:x + h]
label, confidence = face_recognizer.predict(roi_gray)
#print("label:",label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
if confidence < 72:
print(x1, " ", ctr, " ", nctr, " ", label, " ", confidence,
" ", predicted_name, " ", x2)
fr.put_text(test_img, predicted_name, x, y)
if x1 == label and x2 == predicted_name:
ctr = ctr + 1
if ctr == 10:
messagebox.showinfo('AKKI', 'Operation Successful')
cap.release()
cv2.destroyAllWindows()
return ctr
if x1 != label or x2 != predicted_name:
nctr = nctr + 1
if nctr == 10:
messagebox.showinfo('Akki', 'Operation unsucessful')
cap.release()
cv2.destroyAllWindows()
return 0
cv2.imshow('Cyber Ninja System', test_img)
k = cv2.waitKey(30) & 0xff
if k == 27 or ctr == 20:
break
cap.release()
cv2.destroyAllWindows()
return
def attendance():
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read('trainingData.yml') #Load saved training data
students = Student.query.all()
name = Counter()
i = 0
for ele in students:
ele.status = 0
name[i] = ele
i += 1
db.session.commit()
cap = cv2.VideoCapture(0)
start = time.time()
while i and name:
ret, test_img = cap.read(
) # captures frame and returns boolean value and captured image
faces_detected, gray_img = fr.faceDetection(test_img)
for (x, y, w, h) in faces_detected:
cv2.rectangle(test_img, (x, y), (x + w, y + h), (255, 0, 0),
thickness=7)
resized_img = cv2.resize(test_img, (1000, 700))
cv2.imshow('face detection Tutorial ', resized_img)
cv2.waitKey(10)
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + w, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) #predicting the label of given image
print("confidence:" + str(confidence))
print("label:" + str(label))
fr.draw_rect(test_img, face)
a = Student.query.filter_by(roll=label).first()
if confidence < 39 and a.status == 0:
a.status = 1
db.session.commit()
#winsound.Beep(2500, 1200)
duration = 1 # seconds
freq = 440 # Hz
os.system('play -nq -t alsa synth {} sine {}'.format(
duration, freq))
i -= 1
fr.put_text(test_img, str(a), x, y)
end = time.time()
if cv2.waitKey(30) & 0xFF == ord('q'):
break
if int(end - start) >= 180:
break
cap.release()
cv2.destroyAllWindows()
return redirect(url_for('result'))
def train():
# read time table and class names
read_tt.read_tt_and_names()
# making excel template
excel_template.make_template()
# This module takes images stored in disk and performs face recognition
test_img_name = str(input("image name : "))
test_img = cv2.imread('TestImages/' + test_img_name +
'.jpg') # test_img path
# detect all the faces in image
faces_detected, gray_img = fr.faceDetection(test_img)
print("faces_detected:", faces_detected)
# counter for keeping face count
face_count = len(faces_detected)
print("face_count:", face_count)
faces, faceID = fr.labels_for_training_data('trainingImages')
face_recognizer = fr.train_classifier(faces, faceID)
# have to store our trained data so we can use it later without going through the training process again
face_recognizer.write('trainingData.yml')
# use this .yml file in future to avoid training time
# creating dictionary containing names for each label
name = np.load("names.npy", allow_pickle=True, fix_imports=True)
name = name.item()
# id of students present in class
present = []
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) # predicting the label of given image
present.append(label)
fr.draw_rect(test_img, face) # drawing rectangle on face
predicted_name = name[label]
fr.put_text(test_img, predicted_name, x,
y) # printing name of the person
# # print ids of present students
# present.sort()
print(present)
# cv2.imshow("test image", resized_img)
cv2.waitKey(0) # Waits indefinitely until a key is pressed
cv2.destroyAllWindows()
# train()
def test():
#This module captures images via webcam and performs face recognition
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read('trainingData.yml') #Load saved training data
list = Student.query.all()
name = {}
i = 1
for ele in list:
ele.status = False
name[i] = ele
i = i + 1
cap = cv2.VideoCapture(0)
while True:
ret, test_img = cap.read(
) # captures frame and returns boolean value and captured image
faces_detected, gray_img = fr.faceDetection(test_img)
for (x, y, w, h) in faces_detected:
cv2.rectangle(test_img, (x, y), (x + w, y + h), (255, 0, 0),
thickness=7)
resized_img = cv2.resize(test_img, (1000, 700))
#cv2.imshow('face detection Tutorial ',resized_img)
cv2.waitKey(10)
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + w, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) #predicting the label of given image
print("confidence:", confidence)
print("label:", label)
fr.draw_rect(test_img, face)
predicted_name = str(name[label])
if confidence < 37: #If confidence less than 37 then don't print predicted face text on screen
detected = Student.query.filter_by(Roll=predicted_name)
detected.status = True
db.session.commit()
fr.put_text(test_img, predicted_name, x, y)
resized_img = cv2.resize(test_img, (1000, 700))
cv2.imshow('face recognition tutorial ', resized_img)
if cv2.waitKey(10) == ord('q'): #wait until 'q' key is pressed
break
cap.release()
cv2.destroyAllWindows()
def find_face(imageNames):
haar_file = cv2.CascadeClassifier('haarcascade_frontalface_alt.xml')
test_img = cv2.imread("dest_folder/"+imageNames)
test_img = cv2.resize(test_img, (2000,2000))
test_img = detect_faces(haar_file,test_img)
#cv2.imshow("bikas",test_img)
#name={0:"Bikas",1:"Bishal"}
name={}
db = MySQLdb.connect('localhost', 'root','','facetag_db')
cur = db.cursor()
cur.execute("SELECT * FROM login_tbl")
for row in cur.fetchall():
#imageName=row[2]
uidd=int(row[0])
uidd=uidd-1
name.update({uidd:row[3]})
faces_detected,gray_img=fr.faceDetection(test_img)
print("faces_detected:",faces_detected)
face_recognizer=cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read('trainingData.yml')
for face in faces_detected:
(x,y,w,h)=face
roi_gray=gray_img[y:y+h,x:x+h]
label,confidence=face_recognizer.predict(roi_gray)#predicting the label of given image
print("confidence:",confidence)
print("label:",label)
fr.draw_rect(test_img,face)
predicted_name=name[label]
uidds=label+1
if(userId !=uidds):
if(confidence>47):#If confidence less than 37 then don't print predicted face text on screen
continue
db = MySQLdb.connect('localhost', 'root','','facetag_db')
cur = db.cursor()
sql = "INSERT INTO tagimg_tbl(id,img_id,uid,date,status)VALUES (null,'"+str(image_ids)+"','"+str(label+1)+"', '2020-04-10','0')"
try:
cur.execute(sql)
db.commit()
except pymysql.Error:
db.rollback()
fr.put_text(test_img,predicted_name,x,y)
resized_img=cv2.resize(test_img,(700,700))
cv2.imshow("face detection",resized_img)
cv2.waitKey(0)
cv2.destroyAllWindows()
def b():
import os
import cv2
import numpy as np
import faceRecognition as fr
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read("D:\ML PROJECT\TrainingData.yml")
name = {0: "Alex", 1: "Ashwini", 2: "Amanda"}
cap = cv2.VideoCapture(0)
while True:
ret, test_img = cap.read()
face_detected, gray_img = fr.faceDetection(test_img)
for (x, y, w, h) in face_detected:
cv2.rectangle(test_img, (x, y), (x + w, y + h), (0, 255, 0),
thickness=5)
resized_img = cv2.resize(test_img, (1000, 700))
cv2.imshow("Face Detection", resized_img)
cv2.waitKey(10)
for face in face_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(roi_gray)
print("Confidence:", confidence)
print("Label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
if label == 1:
messagebox.showinfo("Authentication", "face detected")
fr.put_text(test_img, predicted_name, x, y)
resized_img = cv2.resize(test_img, (1000, 700))
cv2.imshow("Face detection", resized_img)
if cv2.waitKey(10) == ord('q'):
break
cap.release()
cv2.destroyAllWindows()
def get_frame(self):
if self.vid.isOpened():
ret, frame = self.vid.read()
predicted_name = 'Take attandence'
label = 0
if ret:
# Return a boolean success flag and the current frame converted to BGR
faces_detected, gray_img = fr.faceDetection(frame)
for (x, y, w, h) in faces_detected:
cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0),
thickness=7)
resized_img = cv2.resize(frame, (1000, 700))
#print(faces_detected)
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + w, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) #predicting the label of given image
if confidence < 39:
print("confidence:", confidence)
print("label:", label)
fr.draw_rect(frame, face)
predicted_name = name[label]
print(predicted_name)
fr.put_text(frame, predicted_name, x, y)
break
return (ret, cv2.cvtColor(frame, cv2.COLOR_BGR2RGB),
predicted_name, label)
else:
return (ret, None, 'Take attandence', 0)
else:
return (ret, None, 'Take attandence', 0)
def a():
import cv2
import os
import numpy as np
import faceRecognition as fr
test_img = cv2.imread("D:\ML PROJECT\Testimg\Ma.jpg")
faces_detected, gray_img = fr.faceDetection(test_img)
print("Face Detected:", faces_detected)
'''for (x,y,w,h) in faces_detected:
cv2.rectangle(test_img,(x,y),(x+w,y+h),(0,255,0),thickness=5)
resized_img = cv2.resize(test_img,(1000,700))
cv2.imshow("Face detection",resized_img)
cv2.waitKey(0)
cv2.destroyAllWindows()'''
# faces,faceID = fr.labels_for_training_data("D:\ML PROJECT\TrainingData")
# face_recognizer = fr.train_classifier(faces,faceID)
# face_recognizer.save('trainingData.yml')
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read("D:\ML PROJECT\TrainingData.yml")
name = {0: "Alex", 1: "Ashwini", 2: "Amanda"}
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(roi_gray)
print("Confidence:", confidence)
print("Label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
if confidence > 37:
continue
fr.put_text(test_img, predicted_name, x, y)
resized_img = cv2.resize(test_img, (800, 600))
cv2.imshow("Face detection", resized_img)
cv2.waitKey(0)
cv2.destroyAllWindows()
def dete():
def audio():
my_text = "user found"
language = 'en'
myobj = gTTS(text=my_text, lang=language, slow=False)
myobj.save("welcome.mp3")
os.system("welcome.mp3")
test_img = cv2.imread('TestImages/divyanshu.jpg')
faces_detected, gray_img = fr.faceDetection(test_img)
print("faces_detected:", faces_detected)
faces, faceID = fr.labels_for_training_data('trainingImages')
face_recognizer = fr.train_classifier(faces, faceID)
face_recognizer.write('trainingData.yml')
name = {0: "chirag", 1: "divyanshu"}
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(roi_gray)
print("confidence:", confidence)
print("label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
if (confidence > 37):
continue
fr.put_text(test_img, predicted_name, x, y)
audio()
resized_img = cv2.resize(test_img, (1000, 1000))
cv2.imshow("face dtecetion ", resized_img)
cv2.waitKey(0)
cv2.destroyAllWindows
def poison(load_img):
test_img = imagePoisoning()
faces_detected, gray_img = fr.faceDetection(load_img)
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read('trainingData.yml')
name = {
0: "Priyanka",
1: "Kangana",
} #creating dictionary containing names for each label
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) #predicting the label of given image
print("confidence:", confidence)
print("label:", label)
fr.draw_rect(load_img, face)
predicted_name = name[label]
if (
confidence < 100
): #If confidence more than 100 then don't print predicted face text on screen
fr.put_text(load_img, predicted_name, x, y)
else:
poison(test_img)
name = {0 : "Mask on", 1:"Without mask"}
cap=cv2.VideoCapture(0)
while True:
ret,test_img=cap.read()# captures frame and returns boolean value and captured image
faces_detected,gray_img=fr.faceDetection(test_img)
for face in faces_detected:
(x,y,w,h)=face
roi_gray=gray_img[y:y+w, x:x+h]
label,confidence=face_recognizer.predict(roi_gray)#predicting the label of given image
print("confidence:",confidence)
print("label:",label)
if label == 0:
fr.draw_rect(test_img, face, 0)
elif label == 1:
fr.draw_rect(test_img, face, 1)
predicted_name=name[label]
#if confidence < 39:#If confidence less than 37 then don't print predicted face text on screen
fr.put_text(test_img,predicted_name,x,y)
for (x,y,w,h) in faces_detected:
if label == '0':
cv2.rectangle(test_img,(x,y),(x+w,y+h),(0,0,255),thickness=1)
elif label == '1':
cv2.rectangle(test_img,(x,y),(x+w,y+h),(0,0,255),thickness=2)
def AuthorizationFunction1(self, Flag):
self.WebCamValue = 0
if Flag == 1:
self.tabSetup.setEnabled(False)
self.tabAuthOut.setEnabled(False)
else:
self.tabSetup.setEnabled(False)
self.tabAuthIn.setEnabled(False)
# This module captures images via webcam and performs face recognition
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read('trainingData.yml') # Load saved training data
session = session_factory()
employee_dao = Employee_Dao()
employees = employee_dao.query_records(session)
name = {}
for employee in employees:
name[employee.id] = employee.name
# print('Names in DB : ', name)
counter = 1
flag = 0
global predicted_name, employeeId
global test_img
global cap
cap = cv2.VideoCapture(0)
while True:
ret, test_img = cap.read() # captures frame and returns boolean value and captured image
self.displayImage_In(test_img, Flag)
faces_detected, gray_img = fr.faceDetection(test_img)
for (x, y, w, h) in faces_detected:
cv2.rectangle(test_img, (x, y), (x + w, y + h), (255, 0, 0), thickness=5)
cv2.waitKey(10)
if self.WebCamValue == 1:
self.WebCamValue = 2
cap.release()
if Flag == 1:
self.lblAuthWebcamScreen.clear()
else:
self.lblAuthWebcamScreenOut.clear()
break
for face in faces_detected:
if counter < 31:
(x, y, w, h) = face
roi_gray = gray_img[y:y + w, x:x + h]
label, confidence = face_recognizer.predict(roi_gray) # predicting the label of given image
# print("confidence:", confidence)
# print("label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
employeeId = label
if confidence < 39: # If confidence less than 37 then don't print predicted face text on screen
# fr.put_text(test_img, predicted_name, x, y)
# fr.put_text(test_img, str(confidence), x, y + h + 50)
flag = 1
counter += 1
elif confidence > 39:
counter += 1
elif flag == 1:
cap.release()
flag = 0
counter = 1
if Flag == 1:
self.InInsertion(current_time, DateToday)
else:
self.OutInsertion(current_time, DateToday)
granted = QMessageBox()
granted.setWindowTitle('Granted !')
granted.setText('Authentication successful. Access granted.'
' Welcome : ' + predicted_name)
granted.setIcon(QMessageBox.Information)
granted.setStandardButtons(QMessageBox.Ok)
granted.setDefaultButton(QMessageBox.Ok)
xy = granted.exec_()
if xy == QMessageBox.Ok:
cap = cv2.VideoCapture(0)
elif flag == 0:
cap.release()
counter = 1
granted = QMessageBox()
granted.setWindowTitle('Denied !')
granted.setText('Authentication unsuccessful. Access denied.')
granted.setIcon(QMessageBox.Question)
granted.setStandardButtons(QMessageBox.Ok)
granted.setDefaultButton(QMessageBox.Ok)
xy = granted.exec_()
if xy == QMessageBox.Ok:
cap = cv2.VideoCapture(0)
) # captures frame and returns boolean value and captured image
faces_detected, gray_img = fr.faceDetection(test_img)
for (x, y, w, h) in faces_detected:
cv2.rectangle(test_img, (x, y), (x + w, y + h), (255, 0, 0),
thickness=7)
#resized_img = cv2.resize(test_img, (1000, 700))
#cv2.imshow('face detection Tutorial ',resized_img)
#cv2.waitKey(10)
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + w, x:x + h]
label, confidence = face_recognizer.predict(
roi_gray) #predicting the label of given image
print("confidence:", confidence)
print("label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
if confidence > 55: #If confidence less than 50 then don't print predicted face text on screen
fr.put_text(test_img, predicted_name, x, y)
resized_img = cv2.resize(test_img, (600, 570))
cv2.imshow('FR', resized_img)
if cv2.waitKey(10) == ord('q'): #wait until 'q' key is pressed
break
cap.release()
cv2.destroyAllWindows
def main():
## faces,faceID = fr.labels_for_training_data("training")
## face_recognizer = fr.train_classifier(faces,faceID)
## face_recognizer.save("trainingData.yml")
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
face_recognizer.read("trainingData.yml")
name = {
0: "Fariha",
1: "Inara",
2: "Arowa",
3: "Ankon",
4: "Farhan",
5: "Minhaz",
6: "Afifa",
7: "Karishma",
8: "Nafisa"
}
gender = {
0: "Female",
1: "Male",
2: "Female",
3: "Male",
4: "Male",
5: "Male",
6: "Female",
7: "Female",
8: "Female"
}
relation = {
0: "Sister",
1: "Cousin",
2: "Aunt",
3: "Uncle",
4: "Nephew",
5: "Child",
6: "Enemy",
7: "Mother",
8: "Daughter"
}
prof = {
0: "Student",
1: "MUA",
2: "Student",
3: "Student",
4: "Student",
5: "Student",
6: "Villain",
7: "Student",
8: "Student"
}
cap = cv2.VideoCapture(0)
while True:
ret, test_img = cap.read()
faces_detected, gray_img = fr.faceDetection(test_img)
for face in faces_detected:
(x, y, w, h) = face
roi_gray = gray_img[y:y + h, x:x + h]
label, confidence = face_recognizer.predict(roi_gray)
print("confidence:", confidence)
print("label:", label)
fr.draw_rect(test_img, face)
predicted_name = name[label]
predicted_relation = relation[label]
predicted_prof = prof[label]
if (confidence < 50):
if (cv2.waitKey(1) == ord('s')):
if (label == 0):
winsound.PlaySound("Fariha.wav", winsound.SND_ASYNC)
elif (label == 1):
winsound.PlaySound("Inara.wav", winsound.SND_ASYNC)
elif (label == 2):
winsound.PlaySound("Arowa.wav", winsound.SND_ASYNC)
elif (label == 3):
winsound.PlaySound("Ankon.wav", winsound.SND_ASYNC)
elif (label == 4):
winsound.PlaySound("Farhan.wav", winsound.SND_ASYNC)
elif (label == 5):
winsound.PlaySound("Minhaz.wav", winsound.SND_ASYNC)
elif (label == 6):
winsound.PlaySound("Afifa.wav", winsound.SND_ASYNC)
elif (label == 7):
winsound.PlaySound("Karishma.wav", winsound.SND_ASYNC)
elif (label == 8):
winsound.PlaySound("Nafisa.wav", winsound.SND_ASYNC)
fr.put_text(test_img, predicted_name, x, y)
fr.put_text2(test_img, predicted_relation, x, y)
fr.put_text2(test_img, predicted_prof, x, y + h - 50)
if (gender[label] == "Female"):
s_img = cv2.imread("pin2.png")
elif (gender[label] == "Male"):
s_img = cv2.imread("pin3.png")
r_img = cv2.imread("relation.png")
p_img = cv2.imread("prof.png")
x_offset = x
y_offset = y - 50
test_img[y_offset:y_offset + s_img.shape[0],
x_offset:x_offset + s_img.shape[1]] = s_img
x1 = x + 10
y1 = y + 10
test_img[y1:y1 + r_img.shape[0],
x1:x1 + r_img.shape[1]] = r_img
x1 = x + 10
y1 = y + h - 40
test_img[y1:y1 + p_img.shape[0],
x1:x1 + p_img.shape[1]] = p_img
resized_img = cv2.resize(test_img, (1000, 700))
cv2.imshow("face detection", resized_img)
if cv2.waitKey(1) == ord('q'):
break
cap.release()
cv2.destroyAllWindows
face_recognizer = cv2.face.LBPHFaceRecognizer_create()
# we load TrainedModel.yml which has our trained model stored.
face_recognizer.read('TrainedModel.yml')
label = {
0: "Priyanka",
1: "Kangana"
} # creating dictionary containing labels for each identity
for face in faces_detected_in_image:
(x, y, w, h) = face
# x is left co-ordinate of image, x+w is right, y is bottom, y+h is top
roi_gray = gray_img[y:y + h, x:x + w]
# predicting the identity of given image and uncertainity of prediction
# lower the uncertainity value, higher the confidence and less are chances of match.
identity, uncertainity = face_recognizer.predict(roi_gray)
# print("uncertainity:",uncertainity)
print("identity:", identity)
fr.draw_rect(subjectImage, face)
predicted_label = label[identity]
# If uncertainity more than 60 then don't print predicted face text on screen
# This is to ensure that our model doesnt wrongly identity any face.
if (uncertainity < 65):
fr.label_face(subjectImage, predicted_label, x, y)
display_image = cv2.resize(subjectImage, (1000, 1000))
cv2.imshow("Face Recognition System", display_image)
cv2.waitKey(0) # wait until some key is pressed.
cv2.destroyAllWindows
