Python+Opencv 实现轻量级人脸检测与识别

一、搭配环境

1 在pychram中搭配好opencv环境

2 到opencv官网下载,因为需要用到其中E:\Edgedownload\opencv\sources\data\haarcascades路径的下训练好的分类器(注意路径要求全英文)

Python+Opencv 实现轻量级人脸检测与识别_第1张图片

Python+Opencv 实现轻量级人脸检测与识别_第2张图片

二、代码

1 首先在项目所在文件夹下新建data文件,在data文件下新建jm文件夹用来放置训练所用图片,在data文件下新建trainer文件下,会保存训练得到的文件。

Python+Opencv 实现轻量级人脸检测与识别_第3张图片

Python+Opencv 实现轻量级人脸检测与识别_第4张图片

 

 2 训练数据

import os
import cv2
import sys
from PIL import Image
import numpy as np

def getImageAndLabels(path):
    #存储人脸数据
    facesSamples=[]
    #存储姓名数据
    ids=[]
    #储存图片信息
    path_list = os.listdir(path)

    path_list.sort(key=lambda x: int(x.split('.')[0]))

    imagePaths=[os.path.join(path,f) for f in path_list]
    #加载分类器
    face_detector = cv2.CascadeClassifier('E:/Edgedownload/opencv/sources/data/haarcascades/haarcascade_frontalface_default.xml')
    #打印数组imagePaths
    print('数据排列:',imagePaths)
    #遍历列表中的图片
    for imagePath in imagePaths:
        #打开图片,黑白化
        PIL_img=Image.open(imagePath).convert('L')
        #将图像转换为数组,以黑白深浅
       # PIL_img = cv2.resize(PIL_img, dsize=(400, 400))
        img_numpy=np.array(PIL_img,'uint8')
        #获取图片人脸特征
        faces = face_detector.detectMultiScale(img_numpy)
        #获取每张图片的id和姓名
        id = int(os.path.split(imagePath)[1].split('.')[0])
        #预防无面容照片
        for x,y,w,h in faces:
            ids.append(id)
            facesSamples.append(img_numpy[y:y+h,x:x+w])
        #打印脸部特征和id
        print('fs:', facesSamples)
        print('id:', id)

    return facesSamples,ids

if __name__ == '__main__':
    #图片路径
    path='./data/jm/'
    #获取图像数组和id标签数组和姓名
    faces,ids=getImageAndLabels(path)
    #获取训练对象
    recognizer=cv2.face.LBPHFaceRecognizer_create()
    #recognizer.train(faces,names)#np.array(ids)
    recognizer.train(faces,np.array(ids))
    #保存文件
    recognizer.write('trainer/trainer.yml')
    #save_to_file('names.txt',names)
    #print(ids)

3 识别测试

import cv2
import numpy as np
import os
# coding=utf-8
import urllib
import urllib.request
import hashlib

#加载训练数据集文件
recogizer=cv2.face.LBPHFaceRecognizer_create()
recogizer.read('trainer/trainer.yml')
names = []

#准备识别的图片
def face_detect_demo(img):
    gray=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)#转换为灰度
    face_detector=cv2.CascadeClassifier('E:/Edgedownload/opencv/sources/data/haarcascades/haarcascade_frontalface_default.xml')
    face=face_detector.detectMultiScale(gray)
    for x,y,w,h in face:
        cv2.rectangle(img,(x,y),(x+w,y+h),color=(0,0,255),thickness=2)
        # cv2.circle(img,center=(x+w//2,y+h//2),radius=w//2,color=(0,255,0),thickness=1)
        # 人脸识别
        ids, confidence = recogizer.predict(gray[y:y + h, x:x + w])
        print('标签id:',ids,'置信评分:', confidence)
        if confidence > 80:
            cv2.putText(img, 'unkonw', (x + 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (0, 255, 0), 1)
        else:
            cv2.putText(img,str(names[ids]), (x + 45, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (0, 255, 0), 1)
    cv2.imshow('result',img)
    #print('bug:',ids)

def name():
    path = './data/jm/'
    path_list = os.listdir(path)
    path_list.sort(key=lambda x: int(x.split('.')[0]))
    imagePaths=[os.path.join(path,f) for f in path_list]
    for imagePath in imagePaths:
       name = str(os.path.split(imagePath)[1].split('.',2)[1])
       names.append(name)
    print(names)

if __name__ == '__main__':
    #输出训练样本图片的名称
    name()
    #打开电脑默认摄像头实时检测
    #cap = cv2.VideoCapture(0)
    # while True:
    #     flag,frame = cap.read()
    #     if not flag:
    #         break
    #     face_detect_demo(frame)
    #     if ord('q') == cv2.waitKey(1):
    #         break

    #打开一张图片实现检测
    cap = cv2.imread('test1.jpg')
    face_detect_demo(cap)
    while True:
        if ord('q') == cv2.waitKey(0):
            break
    #释放内存
    cv2.destroyAllWindows()
    #释放摄像头
    cap.release()

4 结果

 

Python+Opencv 实现轻量级人脸检测与识别_第5张图片

 

 

 


你可能感兴趣的:(opencv,python,计算机视觉,人工智能,pycharm)