Learn how to use opencv and python to realize face recognition!

In this article , We'll look at what face recognition is and how it differs from face detection ？ 

Let's first briefly understand the principle of face recognition , Then jump to the coding section ！！ 
At the end of this article , You can also develop a human face recognition program to recognize faces in images ！！！

Overview of face detection  

What if the machine could automatically detect objects in the image without human intervention ？ 

Everyone's face may be slightly different , But after all, it's safe to say , All faces have specific features . At present, there are various face detection algorithms , but Viola-Jones Algorithm is the oldest method still in use today .

Face detection is usually a step towards many face related applications （ Such as face recognition or face verification ） The first step . however , Face detection has very useful applications . One of the most successful applications of face detection may be “ Taking pictures ”. Example ： When you click a friend's photo , The camera with built-in face detection algorithm will detect the position of the face and adjust the focal length accordingly .

Overview of face recognition  

Generally speaking , Face recognition is a method of face recognition or personal identity verification . Various algorithms can be used for face recognition , But their accuracy may vary . In this article, I will discuss with you how to use deep learning for face recognition . Now let's learn how to use deep learning to recognize faces . 

ad locum , We use face embedding , Each face is converted into a vector . The technology of converting face into vector is called depth measurement learning . Let me divide this process into three simple steps , In order to be better 、 It's easier to understand ：

1. Face detection ： 

The first task we perform is to detect the image （ Photo ） Or face in video stream . Now we know the exact coordinates of the face / Location , So we extract this face for further processing . 

 2. feature extraction ： 

Now we see that we have cut the face from the image , Therefore, we extract specific features . ad locum , We will see how to use face embedding to extract these features of human face .

As we know , The neural network takes the face image as the input , And output a vector representing the most important features of the face ！ In machine learning , This vector is nothing more than an embedding , So we call it vector face embedding .

Now how will this help identify different people's faces ？ When we train neural networks , Network learning outputs similar vectors for faces that look similar . Let's consider an example , If I have multiple face images at different time intervals , Obviously, some characteristics may change, but not too much . So in this case , The vectors associated with faces are similar , Or we can say that they are very close in vector space . 

thus , We've seen how this network works , Let's see how to use this network on our own data . ad locum , We pass all the images in the data to this pre trained network , To get the corresponding embeddings and save them in a file for the next step .

3. Face comparison ： 

We save the face embedding of each face in the data in the file , The next step is to identify new images that are not in our data . therefore , The first step is to calculate the face embedding of the image using the same network we used before , Then compare this embedding with the rest we have . If the generated embedding is closer or similar to any other embedding , We can recognize faces .

Understand what is OpenCV 

In the field of artificial intelligence , Computer vision is one of the most interesting and challenging tasks . Computer vision acts as a bridge between computer software and visualization . Computer vision allows computer software to understand and understand the visualization of the surrounding environment . 

Let's look at an example ： According to the shape 、 Color and size determine the fruit . This task is easy for the human brain , But in the computer vision pipeline , First, we need to collect data , Then we perform data processing operations , Then we train and teach the model to understand how to distinguish fruit based fruit The size of the fruit 、 Shape and color .

Now , There are various software packages that can be used to perform machine learning 、 Deep learning and computer vision problems . up to now , Computer vision is the best module to solve such complex problems .OpenCV It's an open source library . It is used by different programming languages such as R、Python Other support . It may run on most platforms , Such as Windows、Linux and macOS.

OpenCV The advantages of ：

1. Open CV It's free. , It's an open source library . 

2. Open CV Fast , Because it uses C/C++ language-written , Compared with other languages  

3. System RAM The less ,OpenCV The better the result. . 

4. Support most operating systems , Such as Windows、Linux、macOS. 

perform

In this section , We will use OpenCV and Python Face recognition is realized . 

First , Let's see what libraries we need and how to install them ： 

1.OpenCV 

2.dlib 

3. Face recognition

OpenCV Is a video and image processing library , For image and video analysis , Such as face detection 、 License plate reading 、 Photo editing 、 Advanced robot vision, etc . 

dlib The library contains our “ Depth measurement learning ” Realization , It is used to construct a face embedding algorithm for the actual recognition process .

face_recognition The library is very easy to use , We'll use it in our code . 

First , In the installation face_recognition Remember to install before dlib library . Use OpenCV The output of any face recognition application will be as follows ：

To install OpenCV、dlib And face recognition , Type the following code snippet at the command prompt .

pip install opencv-pythonconda install -c conda-forge dlibpip install face_recognition
 Copy code 

Now let's do the code ！ 

Extract features from human faces  

First you need a data set , Even create your own dataset . Just make sure all the images are arranged in a folder , Each folder contains only one person's image .

Now? , Save the dataset in the same folder as the file you want to make . This is the code. , I added comments where needed ：

from imutils import paths #imutils includes opencv functions
import face_recognition
import pickle
import cv2
import os

#get paths of each file in folder named Images
#Images here that contains data(folders of various people)
imagePath = list(paths.list_images('Images'))
kEncodings = []
kNames = []

# loop over the image paths
for (i, ip) in enumerate(imagePath):
# extract the person name from the image path
name = ip.split(os.path.sep)[-2]
# load the input image and convert it from BGR
image = cv2.imread(ip)
rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

boxes = face_recognition.face_locations(rgb,model='hog')
# compute the facial embedding for the any face
encodings = face_recognition.face_encodings(rgb, boxes)
# loop over the encodings
for encoding in encodings:
kEncodings.append(encoding)
kNames.append(name)

#save emcodings along with their names in dictionary data
data = {"encodings": kEncodings, "names": kNames}
#use pickle to save data into a file for later use
f = open("face_enc", "wb")
f.write(pickle.dumps(data))#to open file in write mode
f.close()#to close file
 Copy code 

Now we see that we have stored the embedded in a file named “face_enc” In the file of , So we can use them to identify images （ Photo ） Or face in real-time video stream . In the next section, we'll see how to recognize faces from images .

How to recognize the face in the image  

The following script is used to detect and recognize faces in images . I give comments next to the code where needed to help beginners understand the code effectively .

import face_recognition
import imutils #imutils includes opencv functions
import pickle
import time
import cv2
import os

#to find path of xml file containing haarCascade file
cfp = os.path.dirname(cv2.__file__) + "/data/haarcascade_frontalface_alt2.xml"
# load the harcaascade in the cascade classifier
fc = cv2.CascadeClassifier(cfp)
# load the known faces and embeddings saved in last file
data = pickle.loads(open('face_enc', "rb").read())

#Find path to the image you want to detect face and pass it here
image = cv2.imread(Path-to-img)
rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
#convert image to Greyscale for HaarCascade
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
faces = fc.detectMultiScale(gray,
scaleFactor=1.1,
minNeighbors=6,
minSize=(60, 60),
flags=cv2.CASCADE_SCALE_IMAGE)

# the facial embeddings for face in input
encodings = face_recognition.face_encodings(rgb)
names = []
# loop over the facial embeddings incase
# we have multiple embeddings for multiple fcaes
for encoding in encodings:
#Compare encodings with encodings in data["encodings"]
#Matches contain array with boolean values True and False
matches = face_recognition.compare_faces(data["encodings"],
encoding)
#set name =unknown if no encoding matches
name = "Unknown"
# check to see if we have found a match
if True in matches:
#Find positions at which we get True and store them
matchedIdxs = [i for (i, b) in enumerate(matches) if b]
count = {}
# loop over the matched indexes and maintain a count for
# each recognized face face
for i in matchedIdxs:
#Check the names at respective indexes we stored in matchedIdxs
name = data["names"][i]
#increase count for the name we got
count[name] = count.get(name, 0) + 1
#set name which has highest count
name = max(count, key=count.get)
# will update the list of names
names.append(name)

# do loop over the recognized faces
for ((x, y, w, h), name) in zip(faces, names):
# rescale the face coordinates
# draw the predicted face name on the image
cv2.rectangle(image, (x, y), (x + w, y + h), (0, 255, 0), 2)
cv2.putText(image, name, (x, y), cv2.FONT_HERSHEY_SIMPLEX,
0.75, (0, 255, 0), 2)
cv2.imshow("Frame", image)
cv2.waitKey(0)
 Copy code 

Output

Conclusion

Now we have completed the process of face recognition .

Expanding reading

Face recognition technology is a high-precision technology 、 Easy to use 、 High stability 、 Difficult to counterfeit biometric technology , It has an extremely broad market application prospect . In public security 、 The defence 、 customs 、 traffic 、 Finance 、 social security 、 There is a wide range of demand in industries and departments such as medical and other civil safety control .

We TSINGSEE The R & D personnel of Qingxi video are also actively developing face detection recently 、 Face recognition 、 People flow statistics 、 Helmet detection, etc AI technology , And actively integrate into the existing video platform . Typical examples are EasyCVR Video convergence cloud service , have AI Face recognition 、 License plate recognition 、 Voice talk 、 Pan tilt control 、 Audible and visual alarm 、 The ability of monitoring video analysis and data collection , It is widely used in residential areas 、 Intelligent access control of buildings , Suspicious personnel around the perimeter are detected 、 In scenes such as traffic statistics in the scenic spot .