face++人脸检测（三）搜索

搜索要分为三步，先创建faceset，添加图片，再搜索。搜索是在faceset中寻找最像输入图片中人脸的内容。

首先创建faceset：

下面是我的代码：

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 2018/1/2 18:43
# @Author  : He Hangjiang
# @Site    : 
# @File    : CreatFcaeSet.py
# @Software: PyCharm

import requests
from json import JSONDecoder
import cv2

http_url = "https://api-cn.faceplusplus.com/facepp/v3/faceset/create"
key = "2PX56nNhKpVd57QNKc9sQt9ASRczWlIr"
secret = "7gwhn-gVp6J3we_vdj_bIlL1bGpuZdlp"

data = {"api_key": key, "api_secret": secret, 'display_name':'FacesStore',"outer_id":"hhj"}

response = requests.post(http_url, data=data)
print(response)
print(response.text)

# req_con = response.content.decode('utf-8')
# req_dict = JSONDecoder().decode(req_con)
#
# print(req_dict)

第二步 添加图片

需要注意的是，这里的face_tokens就是我们之前人脸检测中得到的face_tokens，每个对应一个人脸

代码如下：

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 2018/1/2 18:55
# @Author  : He Hangjiang
# @Site    : 
# @File    : AddFace.py
# @Software: PyCharm

import requests
from json import JSONDecoder
import cv2

http_url = "https://api-cn.faceplusplus.com/facepp/v3/faceset/addface"
key = "2PX56nNhKpVd57QNKc9sQt9ASRczWlIr"
secret = "7gwhn-gVp6J3we_vdj_bIlL1bGpuZdlp"

data = {"api_key": key, "api_secret": secret, "outer_id": "hhj","face_tokens":'d6eee462bc73858ae8ed7accca1f6632,c06b924cb227711477291c346b2cb40f'}

# filepath = "face2.jpg"

# files = {"image_file": open(filepath, "rb")}
response = requests.post(http_url, data=data)
print(response)
print(response.text)

# req_con = response.content.decode('utf-8')
# req_dict = JSONDecoder().decode(req_con)
#
# print(req_dict)

第三步 搜索
额， 番茄是谁大家不用管。。。

代码如下：

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 2018/1/2 19:12
# @Author  : He Hangjiang
# @Site    : 
# @File    : SearchFace.py
# @Software: PyCharm

import requests
from json import JSONDecoder
import cv2

http_url = "https://api-cn.faceplusplus.com/facepp/v3/search"
key = "2PX56nNhKpVd57QNKc9sQt9ASRczWlIr"
secret = "7gwhn-gVp6J3we_vdj_bIlL1bGpuZdlp"
filepath = "face4.jpg"

data = {"api_key": key, "api_secret": secret, "outer_id": "hhj"}
files = {"image_file": open(filepath, "rb")}
response = requests.post(http_url, data=data, files=files)

req_con = response.content.decode('utf-8')
req_dict = JSONDecoder().decode(req_con)
print(req_dict)

faces_token = ["d6eee462bc73858ae8ed7accca1f6632","5e85caf4c983b9523f17bebd77715d17"]

if req_dict["results"][0]["face_token"] in faces_token and req_dict["results"][0]["confidence"]>=req_dict["thresholds"]["1e-5"]:
    print("是番茄")
else:
    print("不是番茄")

那么关于face++人脸识别的简单内容基本就在这里了，更深入的应用我如果有深入了解会写出来的。