基于树莓派的安保巡逻机器人-（一、快速人脸录入与精准人脸识别）

import cv2 import numpy as np import os import time # 创建LBPH人脸识别器并加载训练好的模型 recognizer = cv2.face.LBPHFaceRecognizer_create() recognizer.read('trainer/trainer.yml') # 加载Haar特征分类器用于人脸检测 cascadePath = "haarcascade_frontalface_default.xml" faceCascade = cv2.CascadeClassifier(cascadePath) font = cv2.FONT_HERSHEY_SIMPLEX # 初始化ID计数器 id = 0 # ID与姓名的对应关系：例如 ==> Marcelo: id=1, 等等 names = ['None', 'ID=1', 'ID=2', 'ID=3', 'Z', 'W'] # 初始化并开始实时视频捕捉 cam = cv2.VideoCapture(0) cam.set(3, 640) # 设置视频宽度 cam.set(4, 480) # 设置视频高度 # 定义识别为人脸的最小窗口大小 minW = 0.1 * cam.get(3) minH = 0.1 * cam.get(4) frame_count, pTime, cTime = 0, 0, 0 while True: ret, img = cam.read() # 从摄像头读取图像 img = cv2.flip(img, 1) # 垂直翻转图像 gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) # 转换为灰度图 # 检测人脸 faces = faceCascade.detectMultiScale( gray, scaleFactor=1.2, minNeighbors=5, minSize=(int(minW), int(minH)), ) # 遍历检测到的人脸 for (x, y, w, h) in faces: cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 0), 2) # 绘制人脸框 id, confidence = recognizer.predict(gray[y:y + h, x:x + w]) # 识别人脸 if confidence < 70: # 置信度小于70 id = names[id] # 获取对应的姓名 confidence = " {0}%".format(round(100 - confidence)) # 计算置信度 else: id = "unknown" # 识别为未知 confidence = " {0}%".format(round(100 - confidence)) # 显示姓名和置信度 cv2.putText(img, str(id), (x + 5, y - 5), font, 1, (255, 255, 255), 2) cv2.putText(img, str(confidence), (x + 5, y + h - 5), font, 1, (255, 255, 0), 1) frame_count += 1 # 帧计数 cTime = time.time() # 当前时间 fps = 1 / (cTime - pTime) # 计算FPS pTime = cTime # 更新上一帧时间 text = "FPS : " + str(int(fps)) # 显示FPS cv2.putText(img, text, (20, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 0, 255), 1) cv2.imshow('camera', img) # 显示摄像头画面 k = cv2.waitKey(10) & 0xff # 按'ESC'键退出视频 if k == 27: break # 清理工作 print("\n [信息] 退出程序并清理资源") cam.release() # 释放摄像头 cv2.destroyAllWindows() # 关闭所有OpenCV窗口