ComplexImageTask Recognition

请求参数

type<string>required

ComplexImageTask

class<string>required

recognition

imagesBase64<array>required

仅图像的Base64编码示例: [ “/9j/4AAQSkZJRgABAQEAAAAAAAD…”]

Task（在 metadata 中）<string>required

可能的值：oocl_rotate_new 等。任务名称（英文）。

oocl_rotate_new

创建任务`oocl_rotate_new`

查看全部价格

POST

https://api.capmonster.cloud/createTask

要求

{
	"clientKey": "API_KEY",
	"task": {
		"type": "ComplexImageTask",
		"class": "recognition",
		"imagesBase64": [
			"{background_base64}",
			"{circle_base64}"
		],
		"metadata": {
			"Task": "oocl_rotate_new"
		}
	}
}

背景图像 (background_base64):

圆形图像 (circle_base64):

回应

{
  "errorId":0,
  "taskId":407533072
}

获取任务结果`oocl_rotate_new`

POST

https://api.capmonster.cloud/getTaskResult

要求

{
  "clientKey":"API_KEY",
  "taskId": 407533072
}

回应圆形需要顺时针旋转的度数。

{
  "errorId":0,
  "status":"ready",
  "errorCode":null,
  "errorDescription":null,
  "solution": 
  {
	  "answer":[130.90909],
	  "metadata":{"AnswerType":"NumericArray"}
  }		   
}

oocl_rotate_double_new

创建任务`oocl_rotate_double_new`

查看全部价格

在请求中我们传递三张图片：背景、环形和圆形。

POST

https://api.capmonster.cloud/createTask

要求

{ 
	"clientKey": "API_KEY",
	"task": {
		"type": "ComplexImageTask",
		"class": "recognition",
		"imagesBase64": [
			"{background_base64}",
			"{ring_base64}",
			"{circle_base64}"
		],
		"metadata": {
			"Task": "oocl_rotate_double_new"
		}
	}
}

背景图像 (background_base64):

环形图像 (ring_base64):

圆形图像 (circle_base64):

回应

{
  "errorId":0,
  "taskId":407533072
}

获取任务结果`oocl_rotate_double_new`

POST

https://api.capmonster.cloud/getTaskResult

要求

{
  "clientKey":"API_KEY",
  "taskId": 407533072
}

回应环形需要逆时针旋转的度数，圆形需要顺时针旋转的度数。

{
  "errorId":0,
  "status":"ready",
  "errorCode":null,
  "errorDescription":null,
  "solution": 
  {
	  "answer":[130.90909],
	  "metadata":{"AnswerType":"NumericArray"}
  }		   
}

betpunch_3x3_rotate

创建任务`betpunch_3x3_rotate`

查看全部价格

在请求中我们传递九张图片。图片必须按以下顺序传递：

POST

https://api.capmonster.cloud/createTask

要求

{ 
	"clientKey": "API_KEY",
	"task": {
		"type": "ComplexImageTask",
		"class": "recognition",
		"imagesBase64": [
			"{image_1_Base64}",
			"{image_2_Base64}",
			"{image_3_Base64}",
			"{image_4_Base64}",
			"{image_5_Base64}",
			"{image_6_Base64}",
			"{image_7_Base64}",
			"{image_8_Base64}",
			"{image_9_Base64}",
		],
		"metadata": {
			"Task": "betpunch_3x3_rotate"
		}
	}
}

回应

{
  "errorId":0,
  "taskId":407533072
}

获取任务结果`betpunch_3x3_rotate`

POST

https://api.capmonster.cloud/getTaskResult

要求

{
  "clientKey":"API_KEY",
  "taskId": 407533072
}

回应 "answer":[X,X,X,X,X,X,X,X,X], 其中 X 是每张图片的整数值，范围为 1 到 4。4 表示图像不需要旋转；1-3 表示图像逆时针旋转的次数。

{
	"errorId":0,
	"status":"ready",
	"errorCode":null,
	"errorDescription":null,
	"solution":
	{
		"answer":[4,4,4,4,4,3,1,2,2],
		"metadata":{"AnswerType":"NumericArray"}
	}
}

bls

创建任务 `bls`

查看全部价格

在请求中，我们发送9张以base64格式编码的图片。我们还在 metadata 中传递所需的值TaskArgument。

更多相关内容，请访问我们的博客

POST

https://api.capmonster.cloud/createTask

要求

{
	"clientKey":{{API_key}},
	"task": 
	{
		"type": "ComplexImageTask",
		"class": "recognition",
		"imagesBase64": [
			"image1_to_base64",
			"image2_to_base64",
			"image3_to_base64",
			"image4_to_base64",
			"image5_to_base64",
			"image6_to_base64",
			"image7_to_base64",
			"image8_to_base64",
			"image9_to_base64"
		],
		"metadata": {
			"Task": "bls_3x3",
			"TaskArgument": "123"
		}
	}
}

示例任务：

发送转换为base64的图片：

对于此示例： "TaskArgument": "546"

回应

{
  "errorId":0,
  "taskId":143998457
}

获取任务结果 `bls`

POST

https://api.capmonster.cloud/getTaskResult

要求

{
  "clientKey":"API_KEY",
  "taskId": 143998457
}

回应一个值的数组，元素为 true 或 false，取决于图片中的数字是否为所需的参数。

{
  "errorId":0,
  "status":"ready",
  "errorCode":null,
  "errorDescription":null,
  "solution": 
  {
	"answer":[true,true,false,false,true,false,false,true,true],
	"metadata":{"AnswerType":"Grid"}
  }
}

shein

创建任务 `shein`

查看全部价格

在请求中，我们传递一张 base64 格式的图片。

POST

https://api.capmonster.cloud/createTask

要求

{
    "clientKey": "API_KEY",
    "task": {
	"type": "ComplexImageTask",
	"class": "recognition",
	"imagesBase64": [
	   "base64"
	],
	"metadata": {
	    "Task": "shein"
        }
    }
}

示例任务：

回应

{
  "errorId":0,
  "taskId":143998457
}

获取任务结果 `shein`

POST

https://api.capmonster.cloud/getTaskResult

要求

{
  "clientKey":"API_KEY",
  "taskId": 143998457
}

回应按特定顺序点击的坐标。

{
"solution":
{
"answer":[{"X":68.99999964,"Y":201.954889},{"X":127.99999783999999,"Y":281.54887104},{"X":181.00000776,"Y":49.894734680000006}],
"metadata":{"AnswerType":"Coordinate"}
},
"cost":0.0003,
"status":"ready",
"errorId":0,
"errorCode":null,
"errorDescription":null
}

baidu

创建任务 `baidu`

查看全部价格

注意！

在解决流程开始时，可能会暂时出现 unsolvable 的结果。这不是错误 —— 验证码将在初始化完成后继续成功解决。

在请求中，我们传递一张 base64 格式的图片。

POST

https://api.capmonster.cloud/createTask

要求

{
	"clientKey": "API_KEY",
	"task": 
	{
		"type": "ComplexImageTask",
		"class": "recognition",
		"imagesBase64": ["base64"],
		"metadata": {
			"Task": "baidu"
		}
	}
}

示例任务：

回应

{
	"errorId":0,
	"taskId":143998457
}

获取任务结果 `baidu`

POST

https://api.capmonster.cloud/getTaskResult

要求

{
	"clientKey":"API_KEY",
	"taskId": 143998457
}

回应顺时针旋转图像所需的角度。

{
	"solution":
	{
		"answer":[297],
		"metadata":{"AnswerType":"NumericArray"}
	},
	"cost":0.0005,
	"status":"ready",
	"errorId":0,
	"errorCode":null,
	"errorDescription":null
}

如何查找任务创建所需的所有参数

手动方式

请在浏览器中访问您的网站，该网站包含验证码功能。
右键点击验证码元素，选择 检查（Inspect）。

base64

网页上的图片可以以URL或Base64编码格式直接呈现。要找到所需的值，右键点击图片，选择检查，仔细查看元素部分或网络请求网格——在那里可以找到URL或编码内容。

base64ComplexImage

自动方法

一种方便的自动化获取所有必要参数的方式。某些参数每次页面加载时都会重新生成，因此需要通过浏览器（普通或无头模式，如使用Playwright）提取它们。由于动态参数值的有效期很短，获取后必须立即解决验证码。

重要提示！

提供的代码片段是用于熟悉参数提取的基本示例。具体实现将取决于验证码页面、其结构及使用的HTML元素/选择器。

JavaScript
Python
C#

显示代码 (Node.js)

import { chromium } from 'playwright';

// 从src中提取base64的函数
async function getBase64FromSrc(elementHandle) {
  const src = await elementHandle.getAttribute('src');
  if (src && src.startsWith('data:image')) {
    return src.split(',')[1];
  }
  return null;
}

// 使用截图获取base64的函数（如果src不是base64格式）
async function elementToBase64(elementHandle) {
  const base64 = await getBase64FromSrc(elementHandle);
  if (base64) {
    return base64; // 如果图片已经是base64格式，直接返回
  }
  const buffer = await elementHandle.screenshot();
  return buffer.toString('base64');
}

// 将元素数组转换为base64字符串数组的函数
async function multipleElementsToBase64(elements) {
  const base64Array = [];
  for (const el of elements) {
    const base64 = await elementToBase64(el);
    base64Array.push(base64);
  }
  return base64Array;
}

// -------------------------------------------------------------
// 为不同类型的验证码准备base64
// -------------------------------------------------------------

// 1. oocl_rotate_new (背景 + 圆圈)
async function prepareOoclRotateNew(page) {
  const background = await page.$('img.background');
  const circle = await page.$('img.circle');

  const backgroundBase64 = await elementToBase64(background);
  const circleBase64 = await elementToBase64(circle);

  return { backgroundBase64, circleBase64 };
}

// 2. oocl_rotate_double_new (背景 + 圆环 + 圆圈)
async function prepareOoclRotateDoubleNew(page) {
  const background = await page.$('img.background');
  const ring = await page.$('img.ring');
  const circle = await page.$('img.circle');

  const backgroundBase64 = await elementToBase64(background);
  const ringBase64 = await elementToBase64(ring);
  const circleBase64 = await elementToBase64(circle);

  return { backgroundBase64, ringBase64, circleBase64 };
}

// 3. betpunch_3x3_rotate (9张小图片)
async function prepareBetpunch3x3(page) {
  const sectors = await page.$$('img.sector');

  const sectorsBase64 = await multipleElementsToBase64(sectors);

  return { sectorsBase64 };
}

// 4. bls (9张小图片用于点击)
async function prepareBls(page) {
  const sectors = await page.$$('img.sector');

  const sectorsBase64 = await multipleElementsToBase64(sectors);

  return { sectorsBase64 };
}

// 5. shein (1张大图片)
async function prepareShein(page) {
  const captchaImage = await page.$('img.captcha');

  const captchaBase64 = await elementToBase64(captchaImage);

  return { captchaBase64 };
}

// Playwright示例
async function run() {
  const browser = await chromium.launch();
  const page = await browser.newPage();
  await page.goto('https://example.com'); // 替换为实际URL

  // 获取oocl_rotate_new的base64示例
  const { backgroundBase64, circleBase64 } = await prepareOoclRotateNew(page);
  console.log('Base64背景:', backgroundBase64);
  console.log('Base64圆圈:', circleBase64);

  await browser.close();
}

run();

显示代码

import base64
import asyncio
from playwright.async_api import async_playwright

# 从src中提取base64的函数
async def get_base64_from_src(element_handle):
    src = await element_handle.get_attribute("src")
    if src and src.startswith("data:image"):
        return src.split(",")[1]
    return None

# 使用截图获取base64的函数（如果src不是base64格式）
async def element_to_base64(element_handle):
    base64_str = await get_base64_from_src(element_handle)
    if base64_str:
        return base64_str  # 如果图片已经是base64格式，直接返回
    screenshot = await element_handle.screenshot()
    return base64.b64encode(screenshot).decode("utf-8")

# 将元素数组转换为base64字符串数组的函数
async def multiple_elements_to_base64(elements):
    base64_list = []
    for el in elements:
        base64_str = await element_to_base64(el)
        base64_list.append(base64_str)
    return base64_list

# -------------------------------------------------------------
# 为不同类型的验证码准备base64
# -------------------------------------------------------------

# 1. oocl_rotate_new (背景 + 圆圈)
async def prepare_oocl_rotate_new(page):
    background = await page.query_selector("img.background")
    circle = await page.query_selector("img.circle")

    background_base64 = await element_to_base64(background)
    circle_base64 = await element_to_base64(circle)

    return {"backgroundBase64": background_base64, "circleBase64": circle_base64}

# 2. oocl_rotate_double_new (背景 + 圆环 + 圆圈)
async def prepare_oocl_rotate_double_new(page):
    background = await page.query_selector("img.background")
    ring = await page.query_selector("img.ring")
    circle = await page.query_selector("img.circle")

    background_base64 = await element_to_base64(background)
    ring_base64 = await element_to_base64(ring)
    circle_base64 = await element_to_base64(circle)

    return {"backgroundBase64": background_base64, "ringBase64": ring_base64, "circleBase64": circle_base64}

# 3. betpunch_3x3_rotate (9张小图片)
async def prepare_betpunch3x3(page):
    sectors = await page.query_selector_all("img.sector")
    sectors_base64 = await multiple_elements_to_base64(sectors)

    return {"sectorsBase64": sectors_base64}

# 4. bls (9张小图片用于点击)
async def prepare_bls(page):
    sectors = await page.query_selector_all("img.sector")
    sectors_base64 = await multiple_elements_to_base64(sectors)

    return {"sectorsBase64": sectors_base64}

# 5. shein (1张大图片)
async def prepare_shein(page):
    captcha_image = await page.query_selector("img.captcha")
    captcha_base64 = await element_to_base64(captcha_image)

    return {"captchaBase64": captcha_base64}

# Playwright示例
async def run():
    async with async_playwright() as p:
        browser = await p.chromium.launch()
        page = await browser.new_page()
        await page.goto("https://example.com")

        # 获取oocl_rotate_new的base64示例
        result = await prepare_oocl_rotate_new(page)
        print(result)

        await browser.close()

if __name__ == "__main__":
    asyncio.run(run())

显示代码

using System;
using System.Collections.Generic;
using System.Threading.Tasks;
using Microsoft.Playwright;

class Program
{
    // 从src中提取base64的函数
    private static async Task<string?> GetBase64FromSrcAsync(IElementHandle elementHandle)
    {
        var src = await elementHandle.GetAttributeAsync("src");
        if (!string.IsNullOrEmpty(src) && src.StartsWith("data:image"))
        {
            return src.Split(',')[1];
        }
        return null;
    }

    // 使用截图获取base64的函数（如果src不是base64格式）
    private static async Task<string> ElementToBase64Async(IElementHandle elementHandle)
    {
        var base64Str = await GetBase64FromSrcAsync(elementHandle);
        if (base64Str != null)
        {
            return base64Str; // 如果图片已经是base64格式，直接返回
        }

        var screenshot = await elementHandle.ScreenshotAsync();
        return Convert.ToBase64String(screenshot);
    }

    // 将元素数组转换为base64字符串数组的函数
    private static async Task<List<string>> MultipleElementsToBase64Async(IEnumerable<IElementHandle> elements)
    {
        var base64List = new List<string>();
        foreach (var el in elements)
        {
            var base64Str = await ElementToBase64Async(el);
            base64List.Add(base64Str);
        }
        return base64List;
    }

    // -------------------------------------------------------------
    // 为不同类型的验证码准备base64
    // -------------------------------------------------------------

    // 1. oocl_rotate_new (背景 + 圆圈)
    private static async Task<Dictionary<string, string>> PrepareOoclRotateNewAsync(IPage page)
    {
        var background = await page.QuerySelectorAsync("img.background");
        var circle = await page.QuerySelectorAsync("img.circle");

        var backgroundBase64 = await ElementToBase64Async(background);
        var circleBase64 = await ElementToBase64Async(circle);

        return new Dictionary<string, string>
        {
            ["backgroundBase64"] = backgroundBase64,
            ["circleBase64"] = circleBase64
        };
    }

    // 2. oocl_rotate_double_new (背景 + 圆环 + 圆圈)
    private static async Task<Dictionary<string, string>> PrepareOoclRotateDoubleNewAsync(IPage page)
    {
        var background = await page.QuerySelectorAsync("img.background");
        var ring = await page.QuerySelectorAsync("img.ring");
        var circle = await page.QuerySelectorAsync("img.circle");

        var backgroundBase64 = await ElementToBase64Async(background);
        var ringBase64 = await ElementToBase64Async(ring);
        var circleBase64 = await ElementToBase64Async(circle);

        return new Dictionary<string, string>
        {
            ["backgroundBase64"] = backgroundBase64,
            ["ringBase64"] = ringBase64,
            ["circleBase64"] = circleBase64
        };
    }

    // 3. betpunch_3x3_rotate (9张小图片)
    private static async Task<Dictionary<string, List<string>>> PrepareBetpunch3x3Async(IPage page)
    {
        var sectors = await page.QuerySelectorAllAsync("img.sector");
        var sectorsBase64 = await MultipleElementsToBase64Async(sectors);

        return new Dictionary<string, List<string>>
        {
            ["sectorsBase64"] = sectorsBase64
        };
    }

    // 4. bls (9张小图片用于点击)
    private static async Task<Dictionary<string, List<string>>> PrepareBlsAsync(IPage page)
    {
        var sectors = await page.QuerySelectorAllAsync("img.sector");
        var sectorsBase64 = await MultipleElementsToBase64Async(sectors);

        return new Dictionary<string, List<string>>
        {
            ["sectorsBase64"] = sectorsBase64
        };
    }

    // 5. shein (1张大图片)
    private static async Task<Dictionary<string, string>> PrepareSheinAsync(IPage page)
    {
        var captchaImage = await page.QuerySelectorAsync("img.captcha");
        var captchaBase64 = await ElementToBase64Async(captchaImage);

        return new Dictionary<string, string>
        {
            ["captchaBase64"] = captchaBase64
        };
    }

    public static async Task Main()
    {
        using var playwright = await Playwright.CreateAsync();
        var browser = await playwright.Chromium.LaunchAsync();
        var page = await browser.NewPageAsync();
        await page.GotoAsync("https://example.com");

        var result = await PrepareOoclRotateNewAsync(page);
        Console.WriteLine($"Background base64 length: {result["backgroundBase64"].Length}");
        Console.WriteLine($"Circle base64 length: {result["circleBase64"].Length}");

        await browser.CloseAsync();
    }
}

在Node.js中搜索oocl参数的示例

import { chromium } from "playwright";

// 代理（可选）
// const proxyHost = "";
// const proxyUsername = "";
// const proxyPassword = "";

async function getBase64FromSrc(elementHandle) {
  const src = await elementHandle.getAttribute("src");
  if (src && src.startsWith("data:image")) {
    return src.split(",")[1];
  }
  return null;
}

async function elementToBase64(elementHandle) {
  const base64 = await getBase64FromSrc(elementHandle);
  if (base64) {
    return base64;
  }
  const buffer = await elementHandle.screenshot();
  return buffer.toString("base64");
}

async function extractVerifyImagesBase64(page) {
  const imgSelectors = [".verify-img-panel img", ".verify-sub-block img"];

  const imageHandles = (
    await Promise.all(imgSelectors.map((selector) => page.$$(selector)))
  ).flat();

  if (imageHandles.length === 0) {
    throw new Error(
      "在verify-img-panel或verify-sub-block中未找到图片。"
    );
  }

  const imagesBase64 = [];
  for (const handle of imageHandles) {
    const base64 = await elementToBase64(handle);
    imagesBase64.push({
      src: await handle.getAttribute("src"),
      base64,
    });
  }

  return imagesBase64;
}

async function run() {
  const browser = await chromium.launch({
    headless: false,
    // proxy: {
    //   server: `http://${proxyHost}`,
    //   username: proxyUsername,
    //   password: proxyPassword,
    // },
  });

  const page = await browser.newPage();
  await page.goto(
    "https://example.com/registration/form?appurl=https://example.com"
  );

  try {
    const images = await extractVerifyImagesBase64(page);
    for (const { src, base64 } of images) {
      console.log("来源:", src);
      console.log("Base64:", base64);
    }
  } catch (err) {
    console.error(err.message);
  }

  await browser.close();
}

run();

使用 SDK 库

JavaScript
C#

// https://github.com/ZennoLab/capmonstercloud-client-js

import { CapMonsterCloudClientFactory, ClientOptions, ComplexImageTaskRecognitionRequest } from '@zennolab_com/capmonstercloud-client';

document.addEventListener('DOMContentLoaded', async () => {
	const cmcClient = CapMonsterCloudClientFactory.Create(new ClientOptions({ clientKey: '<your capmonster.cloud API key>' }));
	console.log(await cmcClient.getBalance());

	const citRecognitionRequest = new ComplexImageTaskRecognitionRequest({
		imagesBase64: ['/9xwee/'],
		metaData: { Task: 'oocl_rotate' },
	});

	console.log(await cmcClient.Solve(citRecognitionRequest));
});

// https://github.com/ZennoLab/capmonstercloud-client-dotnet

using Zennolab.CapMonsterCloud.Requests;
using Zennolab.CapMonsterCloud;

class Program
{
	static async Task MainComplexImageTaskRecognition(string[] args)
	{
		var clientOptions = new ClientOptions
		{
			ClientKey = "your_api_key" // Replace with your CapMonster Cloud API key
		};

		var cmCloudClient = CapMonsterCloudClientFactory.Create(clientOptions);

		var recognitionCITRequest = new RecognitionComplexImageTaskRequest
		{
			ImagesBase64 = new List<string>
			{
				"/9j/4AAQSkZJRgABAQEAAAAAAAD…"
			},
			Metadata = new RecognitionComplexImageTaskRequest.RecognitionMetadata
			{
				Task = "oocl_rotate_new"
			}
		};

		var recognitionCITResult = await cmCloudClient.SolveAsync(recognitionCITRequest);

		if(recognitionCITResult.Solution.Answer.IsGrid)
			Console.WriteLine("Captcha Solution: " + recognitionCITResult.Solution.Answer.GridAnswer);
		else if (recognitionCITResult.Solution.Answer.IsNumeric)
			Console.WriteLine("Captcha Solution: " + recognitionCITResult.Solution.Answer.NumericAnswer);
	}
}

ComplexImageTask Recognition

请求参数​

oocl_rotate_new​

创建任务oocl_rotate_new​

获取任务结果oocl_rotate_new​

oocl_rotate_double_new​

创建任务oocl_rotate_double_new​

获取任务结果oocl_rotate_double_new​

betpunch_3x3_rotate​

创建任务betpunch_3x3_rotate​

获取任务结果betpunch_3x3_rotate​

bls​

创建任务 bls​

获取任务结果 bls​

shein​

创建任务 shein​

获取任务结果 shein​

baidu​

创建任务 baidu​

获取任务结果 baidu​

如何查找任务创建所需的所有参数​

手动方式​

base64​

自动方法​

使用 SDK 库​

请求参数

oocl_rotate_new

创建任务`oocl_rotate_new`

获取任务结果`oocl_rotate_new`

oocl_rotate_double_new

创建任务`oocl_rotate_double_new`

获取任务结果`oocl_rotate_double_new`

betpunch_3x3_rotate

创建任务`betpunch_3x3_rotate`

获取任务结果`betpunch_3x3_rotate`

bls

创建任务 `bls`

获取任务结果 `bls`

shein

创建任务 `shein`

获取任务结果 `shein`

baidu

创建任务 `baidu`

获取任务结果 `baidu`

如何查找任务创建所需的所有参数

手动方式

base64

自动方法

使用 SDK 库