基于opencv和卷积神经网络实现面部表情识别并打分项目python源码(带界面).zip

import os import sys import cv2 from PySide2.QtWidgets import * from PySide2.QtCore import * from PySide2.QtGui import * from PySide2.QtUiTools import QUiLoader import resnet import torch import numpy as np import time import random from playsound import playsound EXPRESSIONS = ['anger', 'disgust', 'fear', 'happy', 'sad', 'surprised', 'normal'] num=0 lst=[] tim=10 for i in range(7): #st='D:/AI/facial_expression_recognition/emoji/'+EXPRESSIONS[i]+'.png' imge=cv2.imread(os.path.split(os.path.realpath(__file__))[0]+'/emoji/'+EXPRESSIONS[i]+'.png') lst.append(imge) class Window(QWidget): def __init__(self): super().__init__() # 从文件中加载UI定义 qfile_stats=QFile(os.path.split(os.path.realpath(__file__))[0]+'/window.ui') qfile_stats.open(QFile.ReadOnly) qfile_stats.close() # 从 UI 定义中动态 创建一个相应的窗口对象 # 注意：里面的控件对象也成为窗口对象的属性了 # 比如 self.ui.button , self.ui.textEdit self.ui = QUiLoader().load(qfile_stats) self.net = torch.load(os.path.split(os.path.realpath(__file__))[0]+'/fer2013_resnet_lr_3.pkl', map_location='cpu') self.face_cascade = cv2.CascadeClassifier(os.path.split(os.path.realpath(__file__))[0]+'/haarcascade_frontalface_default.xml') # 加载人脸特征库 self.ui.pushButton.clicked.connect(self.run) self.ui.pushButton_2.clicked.connect(self.close) self.ui.checkBox.stateChanged.connect(self.check) self.ui.checkBox_2.stateChanged.connect(self.check) self.ui.checkBox_3.stateChanged.connect(self.check) self.ui.show() def run(self): global tim global num print('开始') num+=1 st='第'+str(num)+'轮：' self.ui.textEdit.append(st) self.ui.textEdit.moveCursor(QTextCursor.End) cap = cv2.VideoCapture(0,cv2.CAP_DSHOW) # while (not cap.isOpened()): # v_num+=1 # cap = cv2.VideoCapture(v_num) lstn=[] score=0 for i in range(5): output=9 n=random.randint(0,6) while (n in lstn): n=random.randint(0,6) lstn.append(n) st=str(i+1)+'、请做出'+EXPRESSIONS[n]+'表情' self.ui.textEdit.append(st) self.ui.textEdit.moveCursor(QTextCursor.End) #展示表情 show = cv2.cvtColor(lst[n], cv2.COLOR_BGR2RGB) showImage = QImage(show.data,show.shape[1],show.shape[0],QImage.Format_RGB888) self.ui.label_2.setPixmap(QPixmap.fromImage(showImage).scaled(self.ui.label_2.width(), self.ui.label_2.height())) time_start=time.time() while (True): ret, frame = cap.read() # 读取一帧的图像 if ret: # 保证摄像头有读取到图片时才进行下列操作 frame = cv2.flip(frame,1) faces = self.face_cascade.detectMultiScale(frame, scaleFactor=1.15, minNeighbors=5, minSize=(5, 5)) # 检测人脸 if len(faces) > 0: # 有人脸时再执行下列操作 with torch.no_grad(): for (x, y, w, h) in faces: face = frame[y:y + h, x:x + w, 0] # 裁剪出人脸,注意，这里x和y反过来了 face = cv2.resize(face, (48, 48)) face = np.reshape(face, (1, 1, 48, 48)) # 转成网络满足的格式 input = torch.tensor(face / 255, dtype=torch.float32) output = np.argmax(self.net(input).numpy(), axis=1)[0] # print(EXPRESSIONS[output]) # print(x,y,w,h) # plt.imshow(face,cmap='gray') # plt.pause(0.01) # print(face.shape) cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2) # 用矩形圈出人脸 font = cv2.FONT_HERSHEY_SIMPLEX cv2.putText(frame, EXPRESSIONS[int(output)], (x, y), fontFace=font, fontScale=1.0, color=(0, 0, 255), thickness=2) #cv2.imshow('Face Recognition', frame) # end(if ret) frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB) heigt, width = frame.shape[:2] pixmap = QImage(frame, width, heigt, QImage.Format_RGB888) pixmap = QPixmap.fromImage(pixmap).scaled(self.ui.label.width(), self.ui.label.height()) self.ui.label.setPixmap(pixmap) time_end=time.time() t=time_end-time_start if cv2.waitKey(50) == ord('q') or t>tim or int(output)==n: if int(output)==n: st=EXPRESSIONS[n]+'识别成功，得分+1' playsound(os.path.split(os.path.realpath(__file__))[0]+'/music/yes.mp3') score+=1 self.ui.textEdit.append(st) self.ui.textEdit.moveCursor(QTextCursor.End) if t>tim: st=EXPRESSIONS[n]+'识别失败' playsound(os.path.split(os.path.realpath(__file__))[0]+'/music/no.mp3') self.ui.textEdit.append(st) self.ui.textEdit.moveCursor(QTextCursor.End) break #end(while) cap.release() # 释放摄像头 cv2.destroyAllWindows() st='总得分：'+str(score) self.ui.textEdit.append(st) self.ui.textEdit.moveCursor(QTextCursor.End) def check(self): global tim if self.ui.checkBox.isChecked(): tim=10 if self.ui.checkBox_2.isChecked(): tim=5 if self.ui.checkBox_3.isChecked(): tim=3 def close(self): self.ui.close() if __name__ == '__main__': app = QApplication([]) app.setWindowIcon(QIcon(os.path.split(os.path.realpath(__file__))[0]+'/icon/brain.ico')) window= Window() app.exec_()