Face Recognition

This tutorial is specific to CM5 version; it cannot be used directly in CM4 version.

1. Experiment Objective

To learn how RiderPI can recognize faces after collecting facial image data.

2. Experiment Steps

1. Enter the RiderPI system, terminate the RiderPI main program, and enter "ip (where ip is RiderPI's IP):8888" in your browser. After logging in, enter the password "yahboom".

2. Then log in and navigate to the path cd /home/pi/Rider-pi_class/5.AI Visual Recognition Course/17. Face Recognition/ and run face_recognition_Class.ipynb.
3. Running the Program

3. Experimental Phenomena

1. The screen will display a real-time image. At this time, press the button in the upper right corner of the screen to capture and record the face.
2. Each time a face is captured, the screen will display "ok," but this is not sufficient evidence for subsequent recognition.

3. Press the button in the upper left corner of the screen to perform face recognition. The program will print two results: 1. Successful capture, indicating face recognition can be performed. 2. Capture failed, indicating face recognition cannot be performed and needs to be re-captured.
4. After recognizing a face, the screen will display the sequence number of the captured image.

Acquired:

Unacquired:

5. This program will not save previously registered faces upon restarting; you need to re-register for recognition.

4. Experimental Source Code Analysis

def take_photo():
    global i 
    show_ok = False  # 控制是否显示OK字样 Control whether the word 'OK' is displayed
    ok_display_time = 0  # 记录显示OK的时间 Record the time when OK is displayed
    
    while True:
        frame = picam2.capture_array()
       
        #cv2.imshow('frame', frame)
        image_widget.value = bgr8_to_jpeg(frame)
        
        b, g, r = cv2.split(frame)
        img = cv2.merge((r, g, b))
        imgok = Image.fromarray(img)
        mydisplay.ShowImage(imgok) 
        
        if button.press_d():
            cv2.imwrite(f"./Face_P/{i}.jpg", frame)
            
while (1):
    frame = picam2.capture_array()
    face_locations = face_recognition.face_locations(frame)
    face_encodings = face_recognition.face_encodings(frame, face_locations)
    # 在这个视频帧中循环遍历每个人脸 Loop through each face in this video frame
    for (top, right, bottom, left), face_encoding in zip(
            face_locations, face_encodings):
        # 看看面部是否与已知人脸相匹配。 Check if the face matches a known face.
        for i, v in enumerate(total_face_encoding):
            match = face_recognition.compare_faces(
                [v], face_encoding, tolerance=0.5)
            name = "Unknown"
            if match[0]:
                total_image_name[i] = total_image_name[i].replace(folder_path,"")
                name = total_image_name[i] #str(i+1) 
                name = 'NO.' + name
                break
        # 画出一个框，框住脸 Draw a frame to enclose the face
        cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 2)
        # 画出一个带名字的标签，放在框下 Draw a label with a name and place it under the box
        cv2.rectangle(frame, (left, bottom - 35), (right, bottom), (0, 0, 255),
                      cv2.FILLED)
        font = cv2.FONT_HERSHEY_DUPLEX
        try:
            cv2.putText(frame, name, (left + 6, bottom - 6), font, 0.7,
                    (255, 255, 255), 1)
        except :
            name = "Unknown"
            cv2.putText(frame, name, (left + 6, bottom - 6), font, 0.7,
                    (255, 255, 255), 1)
    # 显示结果图像 Display result image
    image_widget.value = bgr8_to_jpeg(frame)

take_photo(): The main business of this function is to collect faces

while (1): This loop performs recognition based on the collected faces. The recognized faces will display their serial numbers on the screen, while those not recognized will display "Unknown".