调用百度API实现人脸识别

1、代码

from aip import AipFace
import cv2
import time
import base64
from PIL import Image
from io import BytesIO
import pyttsx3
# """ 你的 APPID AK SK """
APP_ID = '1965####'
API_KEY = 'YXL65ekIloykyjrT4kzc####'
SECRET_KEY = 'lFiapBoZ5eBwOFyxMbiwQDmClg1u####'

client = AipFace(APP_ID, API_KEY, SECRET_KEY)

# def frame2base64(frame):
#     img = Image.fromarray(frame) #将每一帧转为Image
#     output_buffer = BytesIO() #创建一个BytesIO
#     img.save(output_buffer, format='JPEG') #写入output_buffer
#     byte_data = output_buffer.getvalue() #在内存中读取
#     image = base64.b64encode(byte_data) #转为BASE64
#     return image #转码成功 返回base64编码

def generate():
    camera = cv2.VideoCapture(0)
    engine = pyttsx3.init()
    try:
        while True:
            engine = pyttsx3.init()
            ret, img = camera.read()
            cv2.imwrite("E://Ana/face.png",img)
            cv2.imshow("调用摄像头", img)
            imageType = "BASE64"
            groupIdList = "1,2,3,4"


            """ 如果有可选参数 """
            options = {}
            options["max_face_num"] = 4
            options["match_threshold"] = 70
            options["quality_control"] = "NORMAL"
            options["liveness_control"] = "NONE"
            # options["user_id"] = "233451"
            options["max_user_num"] = 4

#             """ 带参数调用人脸搜索 """
            with open("E://Ana/face.png", 'rb') as fp:
                imageB = base64.b64encode(fp.read())
            image = str(imageB, 'utf-8')
            """ 调用人脸搜索 """
            result = client.search(image, imageType, groupIdList, options)
            engine.runAndWait()
            print(2)
            if result:
                if not result['result']:
                    continue
                name = result['result']['user_list'][0]['user_id']#获取名字
                score = result['result']['user_list'][0]['score']#获取相似度
                if name == 'cgh_1':
                    if score>80:
                        print(score)
                        print(name)
                        engine.say("华来了")
                elif name == 'yjc_1':
                    if score > 80:
                        print(score)
                        print(name)
                        engine.say("杨来了")
                elif name == 'cjy_1':
                    if score > 80:
                        print(score)
                        print(name)
                        engine.say("言来了")
                elif name == 'hjy_1':
                    if score > 80:
                        print(score)
                        print(name)
                        engine.say("怡来了")
                else:
                    print("匹配失败")
            else:
                continue
    except Exception as e:
        print(e)
    finally:
        # 释放资源
        engine.runAndWait()
        camera.release()
        cv2.destroyAllWindows()
 
generate()

2、实现步骤

2.1、获取百度ID和key

在百度API中可以申请到免费的API账号，每天有免费的几千次人脸识别，下图为我申请的API账号，然后在人脸库中创建需要识别的人脸就行了

2.2、创建客户端

APP_ID = '1965####'
API_KEY = 'YXL65ekIloykyjrT4kzc####'
SECRET_KEY = 'lFiapBoZ5eBwOFyxMbiwQDmClg1u####'
client = AipFace(APP_ID, API_KEY, SECRET_KEY)

2.3、实现人脸识别

主要内容就是参数需要自己修改，然后获取视频图像，然后通过对比看一下，对比出来的名字，然后再看一下分数即可，我在里面加了一个电脑播报的，所以只要识别出来，电脑就会播报，延迟不超过以秒。