验证码：EasyDL 机器学习识别与云码平台一站式识别

目录

EasyDL 机器学习识别（实践：京东商城）

（一）批量获取验证码图片

（二）EasyDL机器学习（百度智能云）

（三）调用EasyDLAPI接口识别验证码

云码平台一站式识别（实战：豆瓣网）

（一）获取验证码图片

（二）验证码识别

（三）动态获取验证码

---

EasyDL 机器学习识别（实践：京东商城）

（一）批量获取验证码图片

from selenium import webdriver
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver import ActionChains
import time
import requestsservice = Service(executable_path="../_resources/chromedriver")
driver = webdriver.Chrome(service=service)
driver.get("https://passport.jd.com/uc/login")select = driver.find_element(By.XPATH, '/html/body/div[2]/div[2]/div[1]/div/div[3]/a')
select.click()username = driver.find_element(By.XPATH, '//*[@id="loginname"]')
ActionChains(driver).pause(0.5).click(username).send_keys("xxxxxxxxxx").parform()password = driver.find_element(By.XPATH, '//*[@id="nloginpwd"]')
ActionChains(driver).pause(0.5).click(password).send_keys("xxxxxxxxxx").parform()submit = driver.find_element(By.XPATH, '//*[@id="loginsubmit"]')
submit.click()for i in range(30):time.sleep(1)img = driver.find_element(By.XPATH, '/html/body/div[4]/div/div/div/div[1]/div[2]/div[1]/img')img.screenshot("./img/{}.png".format(i))time.sleep(0.5)refresh = driver.find_element(By.XPATH, '/html/body/div[4]/div/div/div/div[1]/div[2]/div[2]')refresh.click()time.sleep(3)
driver.quit()

（二）EasyDL机器学习（百度智能云）

（1）EasyDL零门槛AI开发平台-立即使用-物体检测（图像）

（2）创建模型:填写模型名称（自定义例如：myModel）+业务描述（自定义例如：Captcha Recognition)+完成创建

（3）数据总览中的创建数据集：填写数据集名称（自定义例如：myModelData）+完成

（4）数据总览中在上述数据集中：点击导入+无标注信息+导入方式：本地导入+上传图片（刚才代码抓取的所有图片）+确认并返回

（4）导入成功后在线标注：右上角添加标签（自定义例如：Target）后确定+标注每个图片的黑色区域即滑块移动位置

（5）标注好后训练模型：选择模型名称+添加数据集+公有云部署+开始训练+等待训练结束+查看模型效果+申请发布

（6）发布模型：公有云部署+提交申请

（三）调用EasyDLAPI接口识别验证码

（1）在上述模型中点击服务详情+查看API文档+点击EasyDL版控制台+公有云部署：应用列表+点击创建应用+输入应用名称+选择接口EasyDL+立即创建+获得此应用的API KEY和Secret Key

(2)再次来到API文档：导航栏中：模型发布-公有云部署-物体检测API调用文档中的复制代码创建新py文件修改代码如下:（通过main.png来测试）

import json
import base64
import requestsIMAGE_FILEPATH = "./img/main.png"PARAMS = {"threshold": 0.3}# 服务详情 中的 接口地址
MODEL_API_URL = "【您的API地址】"# 调用 API 需要 ACCESS_TOKEN。若已有 ACCESS_TOKEN 则于下方填入该字符串
# 否则，留空 ACCESS_TOKEN，于下方填入 该模型部署的 API_KEY 以及 SECRET_KEY，会自动申请并显示新 ACCESS_TOKEN
# ACCESS_TOKEN = "【您的ACESS_TOKEN】"
ACCESS_TOKEN = None
API_KEY = "【您的API_KEY】"
SECRET_KEY = "【您的SECRET_KEY】"print("1. 读取目标图片 '{}'".format(IMAGE_FILEPATH))
with open(IMAGE_FILEPATH, 'rb') as f:base64_data = base64.b64encode(f.read())base64_str = base64_data.decode('UTF8')
print("将 BASE64 编码后图片的字符串填入 PARAMS 的 'image' 字段")
PARAMS["image"] = base64_strif not ACCESS_TOKEN:print("2. ACCESS_TOKEN 为空，调用鉴权接口获取TOKEN")auth_url = "https://aip.baidubce.com/oauth/2.0/token?grant_type=client_credentials"               "&client_id={}&client_secret={}".format(API_KEY, SECRET_KEY)auth_resp = requests.get(auth_url)auth_resp_json = auth_resp.json()ACCESS_TOKEN = auth_resp_json["access_token"]print("新 ACCESS_TOKEN: {}".format(ACCESS_TOKEN))
else:print("2. 使用已有 ACCESS_TOKEN")print("3. 向模型接口 'MODEL_API_URL' 发送请求")
request_url = "{}?access_token={}".format(MODEL_API_URL, ACCESS_TOKEN)
response = requests.post(url=request_url, json=PARAMS)
response_json = response.json()
response_str = json.dumps(response_json, indent=4, ensure_ascii=False)
print("结果:
{}".format(response_str))

云码平台一站式识别（实战：豆瓣网）

（一）获取验证码图片

from selenium import webdriver
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver import ActionChains
import time
import requestsservice = Service(executable_path="../_resources/chromedriver")
driver = webdriver.Chrome(service=service)
driver.get("https://www.douban.com/")
time.sleep(2)frame = driver.find_element(By.XPATH, '//*[@id="anony-reg-new"]/div/div[1]/iframe')
driver.switch_to.frame(frame)
time.sleep(0.5)select = driver.find_element(By.XPATH, '/html/body/div[1]/div[1]/ul[1]/li[2]')
select.click()username = driver.find_element(By.XPATH, '//*[@id="username"]')
ActionChains(driver).pause(0.5).click(username).send_keys("xxxxxxxxxx").parform()password = driver.find_element(By.XPATH, '//*[@id="password"]')
ActionChains(driver).pause(0.5).click(password).send_keys("xxxxxxxxxx").parform()submit = driver.find_element(By.XPATH, '/html/body/div[1]/div[2]/div[1]/div[5]/a')
submit.click()time.sleep(2)
driver.switch_to.frame("tcaptcha_iframe_dy")img = driver.find_element(By.XPATH, '//*[@id="slideBg"]')
img.screenshot("./img2/captcha.png")time.sleep(3)
driver.quit()

（二）验证码识别

云码-自动验证码识别平台：在线测试：点击单图滑块（截图）-上传刚获取的验证码图片-点击测试

以下为通过代码实现方法：登陆账号成功后点击个人中心-点击开发文档-点击python

修改文档代码如下（新建py文件名称为：ydm.py)：

注意根据个人情况修改token

import json
import timeimport requests
import base64class YdmVerify(object):_custom_url = "http://api.jfbym.com/api/YmServer/customApi"_token = ""_headers = {'Content-Type': 'application/json'}def common_verify(self, image, verify_type="60000"):# 数英汉字类型# 通用数英1-4位 10110# 通用数英5-8位 10111# 通用数英9~11位 10112# 通用数英12位及以上 10113# 通用数英1~6位plus 10103# 定制-数英5位~qcs 9001# 定制-纯数字4位 193# 中文类型# 通用中文字符1~2位 10114# 通用中文字符 3~5位 10115# 通用中文字符6~8位 10116# 通用中文字符9位及以上 10117# 定制-XX西游苦行中文字符 10107# 计算类型# 通用数字计算题 50100# 通用中文计算题 50101# 定制-计算题 cni 452payload = {"image": base64.b64encode(image).decode(),"token": self._token,"type": verify_type}print(payload)resp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def slide_verify(self, slide_image, background_image, verify_type="20101"):# 滑块类型# 通用双图滑块  20111payload = {"slide_image": base64.b64encode(slide_image).decode(),"background_image": base64.b64encode(background_image).decode(),"token": self._token,"type": verify_type}resp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def sin_slide_verify(self, image, verify_type="20110"):# 通用单图滑块(截图)  20110payload = {"image": base64.b64encode(image).decode(),"token": self._token,"type": verify_type}resp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def traffic_slide_verify(self, seed, data, href, verify_type="900010"):# 定制-滑块协议slide_traffic  900010payload = {"seed": seed,"data": data,"href": href,"token": self._token,"type": verify_type}resp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def click_verify(self, image, label_image=None, extra=None, verify_type="30100"):# 通用任意点选1~4个坐标 30009# 通用文字点选1(extra,点选文字逗号隔开,原图) 30100# 定制-文字点选2(extra="click",原图) 30103# 定制-单图文字点选 30102# 定制-图标点选1(原图) 30104# 定制-图标点选2(原图,extra="icon") 30105# 定制-语序点选1(原图,extra="phrase") 30106# 定制-语序点选2(原图) 30107# 定制-空间推理点选1(原图,extra="请点击xxx") 30109# 定制-空间推理点选1(原图,extra="请_点击_小尺寸绿色物体。") 30110# 定制-tx空间点选(extra="请点击侧对着你的字母") 50009# 定制-tt_空间点选 30101# 定制-推理拼图1(原图,extra="交换2个图块") 30108# 定制-xy4九宫格点选(原图,label_image,image) 30008payload = {"image": base64.b64encode(image).decode(),# "label_image": base64.b64encode(label_image).decode(),"token": self._token,"type": verify_type}if extra:payload['extra'] = extraresp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def rotate(self, out_ring_image, inner_circle_image):# 定制-X度单图旋转  90007# payload = {#     "image": base64.b64encode(image).decode(),#     "token": self._token,#     "type": "90007"# }# 定制-Tt双图旋转,2张图,内圈图,外圈图  90004payload = {"out_ring_image": base64.b64encode(out_ring_image).decode(),"inner_circle_image": base64.b64encode(inner_circle_image).decode(),"token": self._token,"type": "90004"}resp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def google_verify(self, googlekey, pageurl, invisible=1, data_s=""):_headers = {'Content-Type': 'application/json'}"""第一步，创建验证码任务:param:return taskId : string 创建成功的任务ID"""url = "http://122.9.52.147/api/YmServer/funnelApi"payload = json.dumps({"token": self._token,# "type": "40011", ## v3"type": "40010",  ## v2"googlekey": googlekey,"enterprise": 1,  ## 是否为企业版"pageurl": pageurl,"invisible": invisible,"data-s": data_s,# 'action':"TEMPLATE" #V3必传})# 发送JSON格式的数据result = requests.request("POST", url, headers=_headers, data=payload).json()print(result)# {'msg': '识别成功', 'code': 10000, 'data': {'code': 0, 'captchaId': '51436618130', 'recordId': '74892'}}captcha_id = result.get('data').get("captchaId")record_id = result.get('data').get("recordId")times = 0while times < 150:try:url = f"http://122.9.52.147/api/YmServer/funnelApiResult"data = {"token": self._token,"captchaId": captcha_id,"recordId": record_id}result = requests.post(url, headers=_headers, json=data).json()print(result)# {'msg': '结果准备中，请稍后再试', 'code': 10009, 'data': []}if result['msg'] == "结果准备中，请稍后再试":time.sleep(5)times += 5continueif result['msg'] == '请求成功' and result['code'] == 10001:print(result['data']['data'])return result['data']['data']# {'msg': '请求成功', 'code': 10001, 'data': {'data': '03AGdBq2611GTOgA2v9HUpMMEUE70p6dwOtYyHJQK4xhdKF0Y8ouSGsFZt647SpJvZ22qinYrm6MYBJGFQxMUIApFfSBN6WTGspk6DmFdQAoWxynObRGV7qNMQOjZ_m4w3_6iRu8SJ3vSUXH_HHuA7wXARJbKEpU4J4R921NfpKdahgeFD8rK1CFYAqLd5fz4l-8_VRmRE83dRSfkgyTN338evQ1doWKJRipZbk4ie-89Ud0KGdOsP4QzG3stRZgj2oaEoMDSAP62vxKGYqtDEqTcwtlgo-ot3rF5SmntaoKGwcKPo0NrekWA5gtj0vqKLU6lY2GcnSci_tgBzBwuH40uvyR1PFu02VK_E44mopJ7FOO4cUukNaLGqypU2YCA8QuaaebOIoCMU7RGqGs_41RYNCG1GSdthiwcwk2hHFbi-TXuICXSwh4Er5mgVW9A3t_9Ndp0eJcyr3HtuJrcA7BtlcgruuQxK5h4Ew4ert4KPH_aQGN9ww5VsUtbSManzUDnUOs7aEdvFk1DOOPmLys-aX20ZFN2CcQcZZSO-7HZpZZt3EDeWWE5S02HFDY8gl3_0xqIts8774Tr4GMVJaddG0NR6pcBFC11FqNcK2a18gM3gaKDy3_2ZMeSU4nj4NWwoAhPjQN2BS8JxX4kKVpX4rD959kc93vczVD3TYD6_4GJahGSpBvM7Y5_GGIdLL8imXde1R35mZnEcFYXQ40zcy3DdJFkk_gzGTVOEb1Q1IZpjMxzCxyGgwjgL9dtDIgst5H5CSZoerX_Lz-DmsBvYIYZdpbPLEMROx9MODImaEw8Cp6M8Xj7_foijiGE9hh-pzJSTlKl3HytiSUyJJ7r1BssrX5C_TFWxl0IXNg8azP8H-ZIOWwnYlMWCS1w9piHdoLg5zACiYIN3Txdlsvi61MuPmzJggJd1_dlyMdAlzb5_zdfweqj0_Ko1ODP378YT7sV7LECgRj5QJU6sF5nlf4m2g5sFypBw9GFAkEE-OaWGYxRJOy2ioU41ggAJIkcza2B_N5AL2KLROtm0-c2MxplM4ZzHxrUv9A24zlgzo3Pz4NONwU_gaOcDB7j1dZKXD8UaoIrZv0BTd8JeojYowm9Usdg7Rt4Fpo_vDLJdrEUfbxVlXieDD9Fr1fu72-d4AduT_J3n-rIhyX4gFav-KfP-qOxqOZsmjXZirsBxZs7042NYeirRYnLv35cxIAJARz03FJmeKViUivwC5mCWw64hjRad9XyyBOP2n8KFOrTXhPskC-WwEfksGtfLxi6VW76FHGvRdwHXzMwVfNqe3P5H_WZUc-vxeTAsTnqZz3WA97lM4MLrX0nTZYgXxCEiS6raSOiEMqcx_Nv7Zxre-abj4LZRbFpH8nx1SEiaOV2Dm-a1iPFEmCs0L4kDtt6VImSVIQaTOAd3KFSo7W_XTvRPsQJOtblrcKyuagztX_Yr0lT0YqN9I9MZAARo7M5OfwSLJW16rdmp4NuRefEvNPNHO2cVh1Xha1qNGuF_QDvWFFmWG0Y6IbRqLmF-Dv8BY4TWyOeVnADJftGQw2QSr8RmbCHryA'}}except Exception as e:print(e)continuedef fun_captcha_verify(self, publickey, pageurl, verify_type="40007"):# 定制类接口-Hcaptcha 40007payload = {"publickey": publickey,"pageurl": pageurl,"token": self._token,"type": verify_type}resp = requests.post(self._custom_url, headers=self._headers, data=json.dumps(payload))print(resp.text)return resp.json()['data']['data']def hcaptcha_verify(self):# 定制类接口-Hcaptcha_headers = {'Content-Type': 'application/json'}_custom_url = "http://api.jfbym.com/api/YmServer/funnelApi"payload = {"sitekey": "","pageurl": "","token": self._token,"type": '50013'}result = requests.post(_custom_url, headers=_headers, data=json.dumps(payload)).json()print(result)captcha_id = result.get('data').get("captchaId")record_id = result.get('data').get("recordId")times = 0while times < 150:try:url = f"http://api.jfbym.com/api/YmServer/funnelApiResult"data = {"token": self._token,"captchaId": captcha_id,"recordId": record_id}result = requests.post(url, headers=_headers, json=data).json()print(result)# {'msg': '结果准备中，请稍后再试', 'code': 10009, 'data': []}if result['msg'] == "结果准备中，请稍后再试":time.sleep(5)times += 5continueif result['msg'] == '请求成功' and result['code'] == 10001:print(result['data']['data'])return result['data']['data']# {'msg': '请求成功', 'code': 10001, 'data': {'data': '03AGdBq2611GTOgA2v9HUpMMEUE70p6dwOtYyHJQK4xhdKF0Y8ouSGsFZt647SpJvZ22qinYrm6MYBJGFQxMUIApFfSBN6WTGspk6DmFdQAoWxynObRGV7qNMQOjZ_m4w3_6iRu8SJ3vSUXH_HHuA7wXARJbKEpU4J4R921NfpKdahgeFD8rK1CFYAqLd5fz4l-8_VRmRE83dRSfkgyTN338evQ1doWKJRipZbk4ie-89Ud0KGdOsP4QzG3stRZgj2oaEoMDSAP62vxKGYqtDEqTcwtlgo-ot3rF5SmntaoKGwcKPo0NrekWA5gtj0vqKLU6lY2GcnSci_tgBzBwuH40uvyR1PFu02VK_E44mopJ7FOO4cUukNaLGqypU2YCA8QuaaebOIoCMU7RGqGs_41RYNCG1GSdthiwcwk2hHFbi-TXuICXSwh4Er5mgVW9A3t_9Ndp0eJcyr3HtuJrcA7BtlcgruuQxK5h4Ew4ert4KPH_aQGN9ww5VsUtbSManzUDnUOs7aEdvFk1DOOPmLys-aX20ZFN2CcQcZZSO-7HZpZZt3EDeWWE5S02HFDY8gl3_0xqIts8774Tr4GMVJaddG0NR6pcBFC11FqNcK2a18gM3gaKDy3_2ZMeSU4nj4NWwoAhPjQN2BS8JxX4kKVpX4rD959kc93vczVD3TYD6_4GJahGSpBvM7Y5_GGIdLL8imXde1R35mZnEcFYXQ40zcy3DdJFkk_gzGTVOEb1Q1IZpjMxzCxyGgwjgL9dtDIgst5H5CSZoerX_Lz-DmsBvYIYZdpbPLEMROx9MODImaEw8Cp6M8Xj7_foijiGE9hh-pzJSTlKl3HytiSUyJJ7r1BssrX5C_TFWxl0IXNg8azP8H-ZIOWwnYlMWCS1w9piHdoLg5zACiYIN3Txdlsvi61MuPmzJggJd1_dlyMdAlzb5_zdfweqj0_Ko1ODP378YT7sV7LECgRj5QJU6sF5nlf4m2g5sFypBw9GFAkEE-OaWGYxRJOy2ioU41ggAJIkcza2B_N5AL2KLROtm0-c2MxplM4ZzHxrUv9A24zlgzo3Pz4NONwU_gaOcDB7j1dZKXD8UaoIrZv0BTd8JeojYowm9Usdg7Rt4Fpo_vDLJdrEUfbxVlXieDD9Fr1fu72-d4AduT_J3n-rIhyX4gFav-KfP-qOxqOZsmjXZirsBxZs7042NYeirRYnLv35cxIAJARz03FJmeKViUivwC5mCWw64hjRad9XyyBOP2n8KFOrTXhPskC-WwEfksGtfLxi6VW76FHGvRdwHXzMwVfNqe3P5H_WZUc-vxeTAsTnqZz3WA97lM4MLrX0nTZYgXxCEiS6raSOiEMqcx_Nv7Zxre-abj4LZRbFpH8nx1SEiaOV2Dm-a1iPFEmCs0L4kDtt6VImSVIQaTOAd3KFSo7W_XTvRPsQJOtblrcKyuagztX_Yr0lT0YqN9I9MZAARo7M5OfwSLJW16rdmp4NuRefEvNPNHO2cVh1Xha1qNGuF_QDvWFFmWG0Y6IbRqLmF-Dv8BY4TWyOeVnADJftGQw2QSr8RmbCHryA'}}except Exception as e:print(e)continue

新建py文件：

from ydm import YdmVerifyY = YdmVerify()
with open('./img2/captcha.png','rb') as f:s = f.read()
Y.sin_slide_verify(image=s)

（三）动态获取验证码

根据一和二结合代码，修改py文件

from selenium import webdriver
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver import ActionChains
import time
import requests
from ydm import YdmVerifyservice = Service(executable_path="../_resources/chromedriver")
driver = webdriver.Chrome(service=service)
driver.get("https://www.douban.com/")
time.sleep(2)frame = driver.find_element(By.XPATH, '//*[@id="anony-reg-new"]/div/div[1]/iframe')
driver.switch_to.frame(frame)
time.sleep(0.5)select = driver.find_element(By.XPATH, '/html/body/div[1]/div[1]/ul[1]/li[2]')
select.click()username = driver.find_element(By.XPATH, '//*[@id="username"]')
ActionChains(driver).pause(0.5).click(username).send_keys("xxxxxxxxxx").parform()password = driver.find_element(By.XPATH, '//*[@id="password"]')
ActionChains(driver).pause(0.5).click(password).send_keys("xxxxxxxxxx").parform()submit = driver.find_element(By.XPATH, '/html/body/div[1]/div[2]/div[1]/div[5]/a')
submit.click()time.sleep(2)
driver.switch_to.frame("tcaptcha_iframe_dy")img = driver.find_element(By.XPATH, '//*[@id="slideBg"]')
img.screenshot("./img2/captcha.png")
time.sleep(1)Y = YdmVerify()
with open('./img2/captcha.png','rb') as f:s = f.read()
Y.sin_slide_verify(image=s)time.sleep(3)
driver.quit()