百度ai人脸检测以及处理{‘error_code‘: 222202}_pic not has face

在进行百度ai人脸检测检测到无人脸的时遇到的：

{‘error_code‘: 222202, ‘error_msg‘: ‘pic not has face‘, ‘log_id‘: 1017599795575, ‘timestamp‘: 1624}

需求：遇到没有人脸的图片数据需要进行跳过或者删除图片处理

一开始使用的是

"""有bug代码"""
if response:
    for i in response.json()["result"]["face_list"]:
        probability = i["face_probability"]
        if probability > 0.8:
            print("人脸的概率高",probability )
            return "same"
 
 
 
"""
遇到没有人脸的图片数据报错
for i in response.json()["result"]["face_list"]:
TypeError: 'NoneType' object is not subscriptable
"""

看到NoneType 还以为是昨天刚遇到的问题，于是就用 is None 来进行解决但是还是没有解决

"""有bug代码代码"""
if response is None:
    print("没有人脸")
 
"""
此时不管是有人脸还是没有人脸都走不进来
"""

最后重回思考了下，不管有没有人脸的图片数据response都是有数据的，所以此时使用 is None 来判断response是不合适的

以下是对response数据进行分析得来的思路过程

""""有人脸的response"""
 
{'error_code': 0, 'error_msg': 'SUCCESS', 'log_id': 5575201050012, 'timestamp': 1624160360, 'cached': 0, 'result': {'face_num': 1, 'face_list': [{'face_token': 'ae01b8fe10afa0c9a536c27c58f62b04', 'location': {'left': 37.94, 'top': 69.26, 'width': 216, 'height': 199, 'rotation': 3}, 'face_probability': 1, 'angle': {'yaw': 0.5, 'pitch': 11.86, 'roll': 1.05}}]}}
 
 
"""没有人脸的response"""
 
{'error_code': 222202, 'error_msg': 'pic not has face', 'log_id': 1010011015575, 'timestamp': 1624160360, 'cached': 0, 'result': None}
 
"""
进行观察可以发现
    有人脸的response[error_msg]='SUCCESS'
    没有人脸的response[error_msg]='pic not has face'
于是我们可以使用这个数值来对图片进行一个过滤
"""

最终的代码：实现了没有人脸的处理，以及人脸概率的再次比对

        if response:
            # print("没有人脸")
            print(response.json())
        if response.json()["error_msg"] == "SUCCESS":
            print("有人脸")
            for i in response.json()["result"]["face_list"]:
                probability = i["face_probability"]
                if probability > 0.8:
                    print("人脸的概率高",probability )
                    return True
                else:
                    print("人脸的概率低", probability)
                    return False
        if response.json()["error_msg"] == "pic not has face":
            print("没有人脸")
            return False

完整的人脸检测代码：

import requests
import json
import base64
class BaiduAI():
    def __init__(self,AK="RxlPVRY7OS0PvxrG4pqZEBuT",SK="dSHHDw7N6nRsbhAd29AMz6XHX0Gzazoz"):
        self.AK = AK
        self.SK = SK
        self.access_tocken = self.get_acess_token()
    def get_acess_token(self):
        host = 'https://aip.baidubce.com/oauth/2.0/token?        grant_type=client_credentials&client_id={}&client_secret={}'.format(self.AK,self.SK)
        response = requests.get(host)
        if response:
            return response.json()["access_token"]
    
    def face_detect(self,img_path):
        '''
        人脸检测与属性分析
        '''
        f = open(img_path, 'rb')
        img = base64.b64encode(f.read())
        f.close()
        print("img_path", img_path)
        request_url = "https://aip.baidubce.com/rest/2.0/face/v3/detect"
        params = json.dumps({"image": str(img, 'utf-8'),
                             "image_type": "BASE64",
                             # "face_field": "gender,age,beauty,race,expression,emotion",
                             "face_type": "LIVE"
                             })
        print("params")
        request_url = request_url + "?access_token=" + self.access_tocken
        headers = {'content-type': 'application/json'}
        response = requests.post(request_url, data=params, headers=headers)
        if response:
            # print("没有人脸")
            print(response.json())
        if response.json()["error_msg"] == "SUCCESS":
            print("有人脸")
            for i in response.json()["result"]["face_list"]:
                probability = i["face_probability"]
                if probability > 0.8:
                    print("人脸的概率高",probability )
                    return True
                else:
                    print("人脸的概率低", probability)
                    return False
        if response.json()["error_msg"] == "pic not has face":
            print("没有人脸")
            return False
 
   
if __name__ == "__main__":
    ai =  BaiduAI()
    ret = ai.face_detect("final_un_mask/unmask.0.jpg")
    ret = ai.face_detect("final_un_mask/unmask.6185.jpg")