百度AI接口（手写文字识别）：https://ai.baidu.com/docs#/OCR-API/9ef46660
实现效果:
步骤一：接入接口
进入上述网站申请账号,然后运行相关代码，获取 access_token 即算完成（由于百度json每30天更新一次，故代码中进行日期更新了的，如何获取accss_token也可见代码）
步骤二：功能介绍:用户输入的图片路径可为网络上的url，也可为本机上的地址，为图省事，图片名称为 ValidateCode.jpg ，由于本人接入的的百度AI接口的手写文字识别，所以一般的验证码应该都可以通过，如果想加入其它功能，那么返回json数据就会有所改变，具体可以见API接口，本人是为了简化理解百度文档介绍
request.urlretrieve(imagepath, 'ValidateCode.jpg') # 下载图片
更新access_json：因为百度API规定：30天更新一次，所以我就把时间提前了。(别乱搞我的密钥呀，我也是为了分享呀QAQ)
2 def accesjson():
flag = 0
fromtime = 1546061002 #起始时间
nowtime = int(time.time()) #2592000恰好为30天，故提前
if nowtime - fromtime > 2000000:
flag = 1
gcontext = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
# client_id 为官网获取的AK， client_secret 为官网获取的SK
host = 'https://aip.baidubce.com/oauth/2.0/token?grant_' \
'type=client_credentials&client_id=Ooj730ZD0Rm7E1dmcPwoZX9s&client_secret=dr5T1icZGqK8ZFyTr4wi2AWbtNKMIsNd'
req = request.Request(host)
response = request.urlopen(req, context=gcontext).read().decode('UTF-8')
result = json.loads(response)
if flag == 1:
return result
else:
return None
图片正式识别：注意，接入功能不一样，放回json数据不一样，具体看返回json就明白了
#返回图片验证码
def vercode():
f = open('ValidateCode.jpg', 'rb')
img = base64.b64encode(f.read())
#不同百度API接口不一样，传递参数不一样，返回json也不一样
host = 'https://aip.baidubce.com/rest/2.0/ocr/v1/handwriting'
headers = {
'Content-Type': 'application/x-www-form-urlencoded'
}
#更换json
if accesjson() == None:
access_token = '24.18591b2e4c97956e0f830db9f66e5373.2592000.1548646630.282335-15301065'
else:
access_token = accesjson()
print('已更换最新json，欢迎继续使用!')
host = host + '?access_token=' + access_token data = {}
data['access_token'] = access_token
data['image'] = img
res = requests.post(url=host, headers=headers, data=data)
req = res.json()
return req['words_result'][0]['words']
完整代码：目前可实现的功能就是网络上面的文字图片识别，或本机图片识别（和之前的抖音图片加载类似。）
拓展：https://ai.qq.com/ (啥B腾讯的API接口，全是PHP，用都知道怎么用，凉凉。)
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2018/12/29 10:48
# @Author : Empirefree
# @File : 17-2-验证码.py
# @Software: PyCharm Community Edition import base64
import requests
from urllib import request
import os
import ssl
import json
import time
import re def IsHttp(imagepath):
if re.search('http', imagepath) != None:
return 1
else:
return 0 #下载验证码
def downloadpic(imagepath):
# imagepath = "http://210.42.38.26:84/jwc_glxt/ValidateCode.aspx"
if IsHttp(imagepath):
request.urlretrieve(imagepath, 'ValidateCode.jpg') # 下载图片 print(os.path.abspath('ValidateCode.jpg')) #百度限制，每30天更换一次access_json
def accesjson():
flag = 0
fromtime = 1546061002 #起始时间
nowtime = int(time.time()) #2592000恰好为30天，故提前
if nowtime - fromtime > 2000000:
flag = 1
gcontext = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
# client_id 为官网获取的AK， client_secret 为官网获取的SK
host = 'https://aip.baidubce.com/oauth/2.0/token?grant_' \
'type=client_credentials&client_id=Ooj730ZD0Rm7E1dmcPwoZX9s&client_secret=dr5T1icZGqK8ZFyTr4wi2AWbtNKMIsNd'
req = request.Request(host)
response = request.urlopen(req, context=gcontext).read().decode('UTF-8')
result = json.loads(response)
if flag == 1:
return result
else:
return None #返回图片验证码
def vercode():
f = open('ValidateCode.jpg', 'rb')
img = base64.b64encode(f.read())
#不同百度API接口不一样，传递参数不一样，返回json也不一样
host = 'https://aip.baidubce.com/rest/2.0/ocr/v1/handwriting'
headers = {
'Content-Type': 'application/x-www-form-urlencoded'
}
#更换json
if accesjson() == None:
access_token = '24.18591b2e4c97956e0f830db9f66e5373.2592000.1548646630.282335-15301065'
else:
access_token = accesjson()
print('已更换最新json，欢迎继续使用!')
host = host + '?access_token=' + access_token data = {}
data['access_token'] = access_token
data['image'] = img
res = requests.post(url=host, headers=headers, data=data)
req = res.json()
return req['words_result'][0]['words'] def checkcode():
imagepath = input('请输入您的图片路径: ')
downloadpic(imagepath)
str = vercode()
return str if __name__ == '__main__': str = checkcode()
print(str)