开放能力
开发平台
行业应用
生态合作
开发与教学
资讯 社区 控制台
技术能力
语音技术
图像技术
文字识别
人脸与人体识别
视频技术
AR与VR
自然语言处理
知识图谱
数据智能
场景方案
部署方案
行业应用
智能教育
智能医疗
智能零售
智能工业
企业服务
智能政务
智能农业
信息服务
智能园区
智能硬件
文字识别

    API文档

    请求说明

    请求示例

    HTTP 方法:POST

    请求URL: https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance

    URL参数:

    参数
    access_token 通过API Key和Secret Key获取的access_token,参考“Access Token获取

    Header如下:

    参数
    Content-Type application/x-www-form-urlencoded

    Body中放置请求参数,参数详情如下:

    请求参数

    参数 是否必选 类型 可选值范围 说明
    image 和 url/pdf_file 三选一 string - 图像数据,base64编码后进行urlencode,需去掉编码头(data:image/jpeg;base64, )
    要求base64编码和urlencode后大小不超过4M,最短边至少15px,最长边最大4096px,支持jpg/jpeg/png/bmp格式
    url 和 image/pdf_file 三选一 string - 图片完整URL,URL长度不超过1024字节,URL对应的图片base64编码后大小不超过4M,最短边至少15px,最长边最大4096px,支持jpg/jpeg/png/bmp格式,当image字段存在时url字段失效
    请注意关闭URL防盗链
    pdf_file 和 image/url 三选一 string - PDF文件,base64编码后进行urlencode,需去掉编码头(data:application/pdf;base64, )
    要求base64编码和urlencode后大小不超过4M
    注:目前仅支持单页PDF识别,如上传的为多页PDF,仅识别第一页
    templateSign string - 模板 ID,自定义模板或预置模板的唯一标示,可用于调用指定的识别模板进行结构化识别,可在「模板管理」页查看并复制使用
    classifierId string - 分类器Id,分类器的唯一标示,可用于调用指定的分类器对传入的图片进行自动分类及识别
    与 templateSign 至少存在一个,如同时存在,则优先级 templateSign > classfierId

    请求代码示例

    提示一:使用示例代码前,请记得替换其中的示例Token、图片地址或Base64信息。

    提示二:部分语言依赖的类或库,请在代码注释中查看下载地址。

    # 请求模板id
    curl -i -k 'https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance?access_token=【调用鉴权接口获取的token】' --data 'templateSign=xxx&image=【图片Base64编码,需UrlEncode】' -H 'Content-Type:application/x-www-form-urlencoded'
    # 请求分类器id
    curl -i -k 'https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance?access_token=【调用鉴权接口获取的token】' --data 'classifierId=xxx&image=【图片Base64编码,需UrlEncode】' -H 'Content-Type:application/x-www-form-urlencoded'
    # -*- coding:UTF-8 -*-
    # -*- encoding: utf-8 -*-
    import base64
    import requests
    
    if sys.version_info.major == 2:
        from urllib import quote
    else:
        from urllib.parse import quote
    
    
    headers = {
            'Content-Type': "application/x-www-form-urlencoded",
            'charset': "utf-8"
        }
    if __name__ == '__main__':
        # 代码中所需的工具包requests
        # 安装方式:pip install requests
        # iocr识别api_url
        recognise_api_url = "https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance"
    
    
        access_token = "your_access_token"
        templateSign = "your_templateSign"
        detectorId = 0
        classifierId = "your_classifier_id"
        # 测试数据路径
        image_path = "your_file_path"
        try:
            with open(image_path, 'rb') as f:
                image_data = f.read()
            
            if sys.version_info.major == 2:
                image_b64 = base64.b64encode(image_data).replace("\r", "")
            else:
                image_b64 = base64.b64encode(image_data).decode().replace("\r", "")
            
            # 请求模板的bodys
            recognise_bodys = "access_token=" + access_token + "&templateSign=" + templateSign + \
                    "&image=" + quote(image_b64.encode("utf8"))
            # 请求分类器的bodys
            classifier_bodys = "access_token=" + access_token + "&classifierId=" + classifierId + \
                    "&image=" + quote(image_b64.encode("utf8"))
            # 请求模板识别
            response = requests.post(recognise_api_url, data=recognise_bodys, headers=headers)
            # 请求分类器识别
            # response = requests.post(recognise_api_url, data=classifier_bodys, headers=headers)
            print response.text
        except Exception as e:
            print e
    package com.baidu.ocr;
    
    import com.baidu.ai.aip.utils.Base64Util;
    import com.baidu.ai.aip.utils.FileUtil;
    import com.baidu.ai.aip.utils.HttpUtil;
    
    
    public class App 
    {
        public static void main(String[] args) throws Exception
        {
            /**
             * 重要提示代码中所需工具类
             * FileUtil,Base64Util,HttpUtil,GsonUtils请从
             * https://ai.baidu.com/file/658A35ABAB2D404FBF903F64D47C1F72
             * https://ai.baidu.com/file/C8D81F3301E24D2892968F09AE1AD6E2
             * https://ai.baidu.com/file/544D677F5D4E4F17B4122FBD60DB82B3
             * https://ai.baidu.com/file/470B3ACCA3FE43788B5A963BF0B625F3
             * 下载
             */
            // iocr识别apiUrl
            String recogniseUrl = "https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance";
    
    
            String filePath = "path\to\your\image.jpg";
            try {
                    byte[] imgData = FileUtil.readFileByBytes(filePath);
                    String imgStr = Base64Util.encode(imgData);
                    // 请求模板参数
                    String recogniseParams = "templateSign=your_template_sign&image=" + URLEncoder.encode(imgStr, "UTF-8");
                    // 请求分类器参数
                    String classifierParams = "classifierId=your_classfier_id&image=" + URLEncoder.encode(imgStr, "UTF-8");
                    
                    
                    String accessToken = "your_access_token";
                    // 请求模板识别
                    String result = HttpUtil.post(recogniseUrl, accessToken, recogniseParams);
                    // 请求分类器识别
                    // String result = HttpUtil.post(recogniseUrl, accessToken, classifierParams);
                    
                    System.out.println(result);
            } catch (Exception e) {
                    e.printStackTrace();
            }
        }
    }
    #include <iostream>
    #include "curl_install/include/curl/curl.h"
    #include <fstream>
    #include "Base64.h"
    #include <cctype>
    #include <iomanip>
    #include <sstream>
    #include <string>
    
    using namespace std;
    // libcurl库下载链接:https://curl.haxx.se/download.html
    // 通用iocr的接口url
    static string ocr_result;
    /**
     * curl发送http请求调用的回调函数,回调函数中对返回的json格式的body进行了解析,解析结果储存在全局的静态变量当中
     * @param 参数定义见libcurl文档
     * @return 返回值定义见libcurl文档
     */
    static size_t callback(void *ptr, size_t size, size_t nmemb, void *stream) {
        // 获取到的body存放在ptr中,先将其转换为string格式
        ocr_result = string((char *) ptr, size * nmemb);
        return size * nmemb;
    }
    
    string url_encode(const std::string &value) {
        ostringstream escaped;
        escaped.fill('0');
        escaped << hex;
        for (std::string::const_iterator i = value.begin(), n = value.end(); i != n; ++i) {
            std::string::value_type c = (*i);
            // Keep alphanumeric and other accepted characters intact
            if (isalnum(c) || c == '-' || c == '_' || c == '.' || c == '~') {
                escaped << c;
                continue;
            }
            // Any other characters are percent-encoded
            escaped << uppercase;
            escaped << '%' << setw(2) << int((unsigned char) c);
            escaped << nouppercase;
        }
    
        return escaped.str();
    }
    // base64.h下载地址
    // https://bj.bcebos.com/v1/iocr-movie/Base64.h
    int iocr_regenize(const std::string &image_base64, const std::string &access_token, const std::string &templateSign, 
        const std::int classifierId, const std::int detectorId) {
        // iocr识别apiUrl
        const static string recognise_api_url = "https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance";
        // 请求模板的参数
        std::string recognise_params = "access_token=" + access_token + "&templateSign=" + templateSign + "&image=" + image_base64;
    
        // 请求分类器的参数
        std::string classifier_params = "access_token=" + access_token 
            + "&classifierId=" + std::to_string(classifierId) 
            + "&image=" + image_base64;
        
        struct curl_slist * headers = NULL;
        headers = curl_slist_append(headers, "Content-Type:application/x-www-form-urlencoded");
        headers = curl_slist_append(headers, "charset:utf-8");
        CURL *curl;
        CURLcode result_code;
        int is_success;
        curl = curl_easy_init();
        if (curl) {
            // 使用libcurl post数据:设定待post的url等
            curl_easy_setopt(curl, CURLOPT_URL, recognise_api_url.c_str());
            curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
            curl_easy_setopt(curl, CURLOPT_POST, 1);
            curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, false);
            curl_easy_setopt(curl,CURLOPT_WRITEFUNCTION,callback);
            // 请求模板
            curl_easy_setopt(curl, CURLOPT_POSTFIELDS, recognise_params.c_str());
            curl_easy_setopt(curl, CURLOPT_POSTFIELDSIZE, recognise_params.size());
            // 请求分类器
            // curl_easy_setopt(curl, CURLOPT_POSTFIELDS, classifier_params.c_str());
            // curl_easy_setopt(curl, CURLOPT_POSTFIELDSIZE, classifier_params.size());
            result_code = curl_easy_perform(curl);
            // http post不成功时错误处理
            if (result_code != CURLE_OK) {
                fprintf(stderr, "curl_easy_perform() failed: %s\n",
                        curl_easy_strerror(result_code));
                is_success = 1;
                return is_success;
            }
            curl_easy_cleanup(curl);
            curl_slist_free_all(headers);
            // 控制台输出识别结果
            cout << ocr_result << endl;
            is_success = 0;
        } else {
            fprintf(stderr, "curl_easy_init() failed.");
            is_success = 1;
        }
        return is_success;
    }
    
    std::string get_image_b64(const std::string &image_path) {
        ifstream f;
        f.open(image_path, ios::in | ios::binary);
        // 定位指针到文件末尾
        f.seekg(0, ios::end);
        int size = f.tellg();
        // 指针重新回到文件头部
        f.seekg(0, ios::beg);
        char* buffer = (char*)malloc(sizeof(char)*size);
        f.read(buffer,size);
        f.close();
        // base64编码
        std::string image_b64 = base64_encode(buffer, size);
        return image_b64;
    }
    
    int main() {
        std::string access_token = "your_access_token";
        std::string image_path = "your_file_path";
        std::string templateSign = "your_templateSign";
        std::int classifierId = "your_classifier_id";
        std::int detectorId = 0;
        // 获取本地图片的base64
        std::string image_b64 = get_image_b64(image_path);
        // urlcode编码
        image_b64 = url_encode(image_b64);
        // 发送post请求
        iocr_regenize(image_b64, access_token, templateSign, classifierId, detectorId);
        return 0;
    }
    <?php
    /**
     * 发起http post请求(REST API), 并获取REST请求的结果
     * @param string $url
     * @param string $param
     * @return - http response body if succeeds, else false.
     */
    function request_post($url = '', $params = '')
    {
        if (empty($url) || empty($params)) {
            return false;
        }
        $headers = array("Content-Type" => "application/x-www-form-urlencoded", "charset" => "utf-8");
        // 初始化curl
        $curl = curl_init();
        curl_setopt($curl, CURLOPT_URL, $url);
        curl_setopt($curl, CURLOPT_HEADER, 0);
        // 要求结果为字符串且输出到屏幕上
        curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
        curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false);
        // post提交方式
        //curl_setopt($curl, CURLOPT_CUSTOMREQUEST, 'POST');
        curl_setopt($curl, CURLOPT_HTTPHEADER, $headers);
        curl_setopt($curl, CURLOPT_POST, 1);
        curl_setopt($curl, CURLOPT_POSTFIELDS, $params);
        // 运行curl
        $data = curl_exec($curl);
        curl_close($curl);
    
        return $data;
    }
    
    function get_image_b64($image_path = '')
    {
        $image_b64 = '';
        if(file_exists($image_path)) {
            $image_data = file_get_contents($image_path);
            print("file exists\n");
            $image_b64_tmp = base64_encode($image_data);
            $image_b64 = urlencode($image_b64_tmp);
            //print($image_b64);
        } else {
            print("file doesn't exists\n");
        }
        return $image_b64;
    }
    // iocr识别api_url
    $recognise_api_url = "https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance";
    
    
    $access_token = "your_access_token";
    $templateSign = "your_templateSign";
    $detectorId = 0;
    $classifierId = "your_classifier_id";
    $image_path = "your_file_path";
    $image_b64 = get_image_b64($image_path);
    // iocr识别bodys
    $recognise_bodys = "access_token=". $access_token. "&templateSign=". $templateSign. "&image=". $image_b64;
    # 分类器请求的bodys
    $classifier_bodys = "access_token=". $access_token. "&classifierId=". $classifierId. "&image=". $image_b64;
    # 请求模板
    $response = request_post($recognise_api_url, $recognise_bodys);
    # 请求分类器
    # $response = request_post($recognise_api_url, $classifier_bodys);
    
    var_dump($response)
    ?>
    using System;
    using System.Net;
    using System.IO;
    using System.Text;
    using System.Web;
    
    namespace iocr_api_demo
    {
        class IocrApiDemo
        {
            static void Main(String[] args)
            {
                PostHttp();
            }
            public static void PostHttp(){
                // fileUtils.cs 类下载地址
                // https://bj.bcebos.com/v1/iocr-movie/FileUtils.cs
                // iocr识别api_url
                String recognise_api_url = "https://aip.baidubce.com/rest/2.0/solution/v1/iocr/recognise/finance";
                
                String access_token = "your_access_token";
                String image_path = "your_file_path";
                String image_b64 = FileUtils.getFileBase64(image_path);
                // iocr按模板id识别的请求bodys
                string templateSign = "yout_templateSign";
                string recognise_bodys = "access_token=" + access_token + "&templateSign=" + templateSign + 
                                "&image=" + HttpUtility.UrlEncode(image_b64);
                // iocr按分类器id识别的请求bodys
                int classifierId = "your_classifier_id";
                String classifier_bodys = "access_token=" + access_token + "&classifierId=" + classifierId + "&image=" + HttpUtility.UrlEncode(image_b64);
    
                HttpWebRequest httpWebRequest = (HttpWebRequest)WebRequest.Create(recognise_api_url);
                httpWebRequest.ContentType = "application/x-www-form-urlencoded";
                httpWebRequest.Method = "POST";
                httpWebRequest.KeepAlive = true;
                try{
                    // 请求模板id
                    byte[] btBodys = Encoding.UTF8.GetBytes(recognise_bodys);
                    // 请求分类器id
                    // byte[] btBodys = Encoding.UTF8.GetBytes(classifier_bodys);
                    httpWebRequest.ContentLength = btBodys.Length;
                    httpWebRequest.GetRequestStream().Write(btBodys, 0, btBodys.Length);
                    HttpWebResponse httpWebResponse = (HttpWebResponse)httpWebRequest.GetResponse();
                    StreamReader streamReader = new StreamReader(httpWebResponse.GetResponseStream());
                    string responseContent = streamReader.ReadToEnd();
                    Console.WriteLine(responseContent);
                    httpWebResponse.Close();
                    streamReader.Close();
                    httpWebRequest.Abort();
                    httpWebResponse.Close();
                } catch (Exception e){
                    Console.Write(e.Message);
                }    
            }
        }
    }

    返回说明

    返回参数

    字段 是否必选 类型 说明
    logid uint64 唯一的log id,用于问题定位
    error_code int 0代表成功,如果有错误码返回可以参考下方错误码列表排查问题
    error_msg string 如果error_code具体的失败信息,可以参考下方错误码列表排查问题
    data jsonObject 识别返回的结果
    + ret jsonArray 识别出来的字段数组,每一个单元里包含以下几个元素
    ++ word_name string isStructured 为 true 时存在,表示字段的名字;如果 isStructured 为 false 时,不存在
    ++ word string 识别的字符串或单字
    ++ location jsonObject 字段在原图上对应的矩形框位置,通过上边距、左边距、宽度、高度表示
    ++ probability jsonObject 字段的置信度,包括平均、最小和方差
    + templateSign string 图片分类结果对应的模板id或指定使用的模版id。templateSign的对应关系为:
    - mixed_receipt:混贴发票,可对粘贴单中的多张不同票据进行检测分类,返回每张发票的类别及识别结果;
    - vat_invoice:增值税发票;
    - taxi:出租车票;
    - roll_ticket:卷票;
    - train_ticket:火车票;
    - quota_invoice:定额发票;
    - travel_itinerary:行程单;
    - car_invoice:汽车票;
    - toll_invoice:通行费发票;
    - printed_invoice:机打发票。
    + scores float 分类置信度,如果指定templateSign,则该值为1
    + isStructured string 表示是否结构化成功,true为成功,false为失败;成功时,返回结构化的识别结果;失败时,如果能识别,按行返回结果,如果不能识别,返回空

    返回示例

    • 使用自定义模板及自定义分类器功能时,返回结果可参考 iOCR通用版-返回示例
    • detectorId = 0时,使用票据检测分类功能,返回结果如下所示:
    {
    	"data": {
    		"ret": [{
    			"ret": [{
    				"word_name": "invoice_code",
    				"word": "113001851019"
    			}, {
    				"word_name": "invoice_rate",
    				"word": "拾元整"
    			}, {
    				"word_name": "invoice_number",
    				"word": "25018254"
    			}],
    			"templateSign": "quota_invoice",
    			"scores": 0.95463621616364,
    			"isStructured": true
    		}, {
    			"ret": [{
    				"word_name": "invoice_code",
    				"word": "113001751019"
    			}, {
    				"word_name": "invoice_rate",
    				"word": "拾元整"
    			}, {
    				"word_name": "invoice_number",
    				"word": "25018226"
    			}],
    			"templateSign": "quota_invoice",
    			"scores": 0.95132255554199,
          "isStructured": true
    		}, {
    			"ret": [{
    				"word_name": "invoice_code",
    				"word": "113001851019"
    			}, {
    				"word_name": "invoice_rate",
    				"word": "拾元整"
    			}, {
    				"word_name": "invoice_number",
    				"word": "25018273"
    			}],
    			"templateSign": "quota_invoice",
    			"scores": 0.95567744970322,
    			"isStructured": true
    		}, {
    			"ret": [{
    				"word_name": "invoice_code",
    				"word": "113001851019"
    			}, {
    				"word_name": "invoice_rate",
    				"word": "拾元整"
    			}, {
    				"word_name": "invoice_number",
    				"word": "25018272"
    			}],
    			"templateSign": "quota_invoice",
    			"scores": 0.9886274933815,
    			"isStructured": true
    		}],
    		"logid": "155601882921432"
    	},
      "error_code": 0,
    	"error_msg": ""
    }
    上一篇
    使用流程
    下一篇
    常见问题