CV：基于keras利用cv2自带两步检测法对《跑男第六季第五期》之如花片段(或调用摄像头)进行实时脸部表情检测

CV之FR：基于Keras框架利用训练好的hdf5模型直接进行人脸识别推理(cv2自带两步检测法)实现对《跑男第六季第五期》之如花视频片段(或调用摄像头)进行实时脸部表情识别

目录

基于Keras框架利用训练好的hdf5模型直接进行人脸识别推理(cv2自带两步检测法)实现对《跑男第六季第五期》之如花视频片段(或调用摄像头)进行实时脸部表情识别

输出结果

设计思路

核心代码

基于Keras框架利用训练好的hdf5模型直接进行人脸识别推理(cv2自带两步检测法)实现对《跑男第六季第五期》之如花视频片段(或调用摄像头)进行实时脸部表情识别 输出结果

视频地址请观看：基于keras利用cv2自带两步检测法对《跑男第六季第五期》之如花片段(或调用摄像头)进行实时脸部表情检测

设计思路

 

核心代码

#CV：基于keras利用cv2自带两步检测法对《跑男第六季第五期》"如花片段"(或调用摄像头)进行实时脸部表情检测——Jason Niu
import cv2
from keras.models import load_model
import numpy as np

detection_model_path = '../trained_models/detection_models/haarcascade_frontalface_default.xml'
emotion_model_path = '../trained_models/emotion_models/fer2013_mini_XCEPTION.102-0.66.hdf5'
emotion_labels = get_labels('fer2013')

frame_window = 10  
emotion_offsets = (20, 40) 

face_detection = load_detection_model(detection_model_path)
emotion_classifier = load_model(emotion_model_path, compile=False)

emotion_target_size = emotion_classifier.input_shape[1:3]

emotion_window = []

cv2.namedWindow('window_frame,by Jason Niu') #摄像头窗口名称
# video_capture = cv2.VideoCapture(0) #函数定义摄像头对象，其参数0表示第一个摄像头，一般就是笔记本的内建摄像头。
video_capture = cv2.VideoCapture("F:\File_Python\Python_example\YOLOv3_use_TF\RunMan5.mp4") 
while True:
    bgr_image = video_capture.read()[1] 
    gray_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2GRAY)
    rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB) 
    faces = detect_faces(face_detection, gray_image) 

    for face_coordinates in faces:  

        x1, x2, y1, y2 = apply_offsets(face_coordinates, emotion_offsets) 
        gray_face = gray_image[y1:y2, x1:x2]  #[坐标参数，尺寸参数]
        try:
            gray_face = cv2.resize(gray_face, (emotion_target_size)) 
        except:
            continue

        gray_face = preprocess_input(gray_face, True) 
        gray_face = np.expand_dims(gray_face, 0) 
        gray_face = np.expand_dims(gray_face, -1)
        emotion_prediction = emotion_classifier.predict(gray_face)

        emotion_probability = np.max(emotion_prediction) 
        emotion_label_arg = np.argmax(emotion_prediction) 
        emotion_text = emotion_labels[emotion_label_arg]  
        emotion_window.append(emotion_text)            

        if len(emotion_window) > frame_window: 
            emotion_window.pop(0)   
        try:
            emotion_mode = mode(emotion_window)
        except:
            continue
        #if条件根据不同表情显示不同颜色
        if emotion_text == 'angry':
            color = emotion_probability * np.asarray((255, 0, 0)) #红色
        elif emotion_text == 'sad':
            color = emotion_probability * np.asarray((0, 0, 255)) #蓝色
        elif emotion_text == 'happy':
            color = emotion_probability * np.asarray((255, 255, 0)) #黄色
        elif emotion_text == 'surprise':
            color = emotion_probability * np.asarray((0, 255, 255)) #青色
        else:
            color = emotion_probability * np.asarray((0, 255, 0))  #绿色

        color = color.astype(int)
        color = color.tolist()

        draw_bounding_box(face_coordinates, rgb_image, color)  
        draw_text(face_coordinates, rgb_image, emotion_mode,
                  color, 0, -45, 1, 4)                        
   
    bgr_image = cv2.cvtColor(rgb_image, cv2.COLOR_RGB2BGR)
    cv2.namedWindow("window_frame,by Jason Niu",0);
    cv2.resizeWindow("window_frame,by Jason Niu", 640, 380);
    cv2.imshow('window_frame,by Jason Niu', bgr_image)     
    if cv2.waitKey(1) & 0xFF == ord('q'):  
        break