阿里云平台的人脸识别接口测试

需要导入的包：

必备模块：import urllib

请求模块：import urllib.request

解析模块：import urllib.parse 可以拆分URL，也可以拼接URL

数据交换格式：import json

需要输入的参数是：

URL地址

method方式：POST

API_KEY：自己申请的号

body：字典，有type；POST的参数：image_1:image的路径+图片

headers：请求头部 {'content-type': 'application/json'}或者('Authorization', 'APPCODE ' + appcode)

涉及到的封装好的方法：

data1 = json.dumps(bodys)：将输入的body字典转化成json格式（Python 对象编码成 JSON 字符串）

data2 = data1.encode('utf-8')：将utf-8的编码格式通过encode编码成python对象是别的Unicode的格式

req1= urllib.request.Request(url, data) ：使用request包里面的Request的方法，对该网址发出POST请求，还要附加data2

request.add_header('Authorization', 'APPCODE ' + appcode)和参数('Content-Type', 'application/json; charset=UTF-8') 请求头

response = urllib.request.urlopen(req1)：将上述请求得到的结果，通过request里面的urlopen的方法，返回一个类文件对象

content = response.read()：对上述获得的类文件对象读取内容

然后就可以打印出读取到的content，看看服务器给你响应了写啥玩意

json.loads

将已编码的 JSON 字符串解码为 Python 对象

代码

前面一个是，输入两张图片，输出两张图片的相似度

后面是，依次遍历lfw人脸识别的数据集，每次输入两张图片，对应输出一个结果

import urllib
import urllib.request
import urllib.parse
import jsonhost = 'http://rlsbbd.market.alicloudapi.com'
path = '/face/verify'
method = 'POST'
appcode = '5axxxxxx39e7xxxxxxxx1fb7'
querys = ''
bodys = {}
url = host + pathbodys = {"type": 0,"image_url_1": "http://vis-www.cs.umass.edu/lfw/images/David_Beckham/David_Beckham_0011.jpg","image_url_2": 'http://vis-www.cs.umass.edu/lfw/images/David_Beckham/David_Beckham_0029.jpg'}
post_data = json.dumps(bodys)
post_data = post_data.encode('utf-8')
request = urllib.request.Request(url, post_data)
request.add_header('Authorization', 'APPCODE ' + appcode)
request.add_header('Content-Type', 'application/json; charset=UTF-8')
response = urllib.request.urlopen(request)
content = response.read()
if (content):print(content)
return content

import urllib
import urllib.request
import urllib.parse
import time
import jsondef compare(image1,image2):host = 'http://rlsbbd.market.alicloudapi.com'path = '/face/verify'method = 'POST'appcode = '5axxxxxxx39exxxxxxxxxxa1fb7'querys = ''bodys = {}url = host + pathbodys = {"type":0,"image_url_1":image1,"image_url_2":image2}post_data = json.dumps(bodys)post_data = post_data.encode('utf-8')request = urllib.request.Request(url, post_data)request.add_header('Authorization', 'APPCODE ' + appcode)request.add_header('Content-Type', 'application/json; charset=UTF-8')response = urllib.request.urlopen(request)content = response.read()if (content):print(content)return contentdef compare_test():lfw_file = open("pairs.txt")res_file1 = open("f1.txt", "a+")res_file2 = open("f2.txt", "a+")count = 0while 1:image_path = "http://vis-www.cs.umass.edu/lfw/images/"line = lfw_file.readline()if not line:breakline = line.strip('\n')images = line.split('\t')if len(images) > 3:register_image = image_path + images[0] + "/" + images[0] + "_"if len(images[1]) < 2:register_image = register_image + "000" + images[1] + ".jpg"elif len(images[1]) < 3:register_image = register_image + "00" + images[1] + ".jpg"elif len(images[1]) < 4:register_image = register_image + "0" + images[1] + ".jpg"else:register_image = register_image + images[1] + ".jpg"verify_image = image_path + images[2] + "/" + images[2] + "_"if len(images[3]) < 2:verify_image = verify_image + "000" + images[3] + ".jpg"elif len(images[3]) < 3:verify_image = verify_image + "00" + images[3] + ".jpg"elif len(images[3]) < 4:verify_image = verify_image + "0" + images[3] + ".jpg"else:verify_image = verify_image + images[3] + ".jpg"else:register_image = image_path + images[0] + "/" + images[0] + "_"if len(images[1]) < 2:register_image = register_image + "000" + images[1] + ".jpg"elif len(images[1]) < 3:register_image = register_image + "00" + images[1] + ".jpg"elif len(images[1]) < 4:register_image = register_image + "0" + images[1] + ".jpg"else:register_image = register_image + images[1] + ".jpg"verify_image = image_path + images[0] + "/" + images[0] + "_"if len(images[2]) < 2:verify_image = verify_image + "000" + images[2] + ".jpg"elif len(images[2]) < 3:verify_image = verify_image + "00" + images[2] + ".jpg"elif len(images[2]) < 4:verify_image = verify_image + "0" + images[2] + ".jpg"else:verify_image = verify_image + images[2] + ".jpg"content = compare(register_image, verify_image)content = str(content)content = content.strip('b\'')content = json.loads(content)if (int(count / 300)) % 2 == 0:flag = 1else:flag = 0if 'confidence' in content.keys():score = content['confidence']else:score = '-1'if flag == 1:res_file1.write(line + '\t' + str(score) + '\t' + str(flag) + '\n')elif flag==0:res_file2.write(line + '\t' + str(score) + '\t' + str(flag) + '\n')count += 1time.sleep(3)lfw_file.close()res_file1.close()res_file2.close()compare_test()

附加：

py文件中的编码

utf-8是 unicode的一种实现方式，unicode、gbk、gb2312是编码字符集.

Python 默认脚本文件都是 ANSCII 编码的，当文件中有非 ANSCII 编码范围内的字符的时候就要使用"编码指示"来修正一个 module 的定义中，如果.py文件中包含中文字符（严格的说是含有非anscii字符），则需要在第一行或第二行指定编码声明：# -*- coding=utf-8 -*- 或者 #coding=utf-8
其他的编码如：gbk、gb2312也可以；否则会出现:

SyntaxError: Non-ASCII character '\xe4' in file test.py on line 3,

python 编码转换函数

unicode 转为 gb2312,utf-8等,使用 encode(encoding)

s = u'中国'
s_gb = s.encode('gb2312')

utf-8,GBK转换为 unicode 使用 unicode(s,encoding) 或者 s.decode(encoding)

s = u'中国'
#s为unicode先转为utf-8
s_utf8 =  s.encode('UTF-8')
assert(s_utf8.decode('utf-8') == s)