OCR发票识别API实现

1. 阿里云OCR发票识别

阿里云OCR发票识别

示例：
接口：https://dgfp.market.alicloudapi.com/ocrservice/invoice
参数：{"img":"图片的base64编码"}
图片转base64网址

结果：

{"angle": 0,"data": {"发票代码": "","发票号码": "23912000000004155751","机打发票代码": "","机打发票号码": "","开票日期": "2023年06月19日","机器编码": "","校验码": "","受票方名称": "百望股份有限公司","受票方税号": "91110108339805094M","受票方地址、电话": "","受票方开户行、账号": "","密码区": "","不含税金额": "350.94","发票税额": "21.05","大写金额": "叁佰柒拾贰圆整","发票金额": "372.00","销售方名称": "华住酒店管理有限公司大连分公司","销售方税号": "91210202071550843F","销售方地址、电话": "","销售方开户行、账号": "","收款人": "","复核人": "","开票人": "辛梦娣","备注": "","标题": "电子发票(增值税专用发票)","联次": "","发票类型": "数电专用发票","特殊标识信息": "","发票详单": [{"货物或应税劳务、服务名称": "*住宿服务*住宿费","规格型号": "","单位": "天","数量": "1","单价": "350.943396226415","金额": "350.94","税率": "6%","税额": "21.06"}],"发票代码解析": ""},"height": 448,"orgHeight": 448,"orgWidth": 660,"prism_keyValueInfo": [{"key": "发票代码","keyProb": 100,"value": "","valueProb": 100},{"key": "发票号码","keyProb": 100,"value": "23912000000004155751","valuePos": [{"x": 516,"y": 18},{"x": 616,"y": 19},{"x": 616,"y": 30},{"x": 516,"y": 29}],"valueProb": 100},{"key": "机打发票代码","keyProb": 100,"value": "","valueProb": 100},{"key": "机打发票号码","keyProb": 100,"value": "","valueProb": 100},{"key": "开票日期","keyProb": 100,"value": "2023年06月19日","valuePos": [{"x": 517,"y": 38},{"x": 588,"y": 38},{"x": 588,"y": 50},{"x": 517,"y": 50}],"valueProb": 100},{"key": "机器编码","keyProb": 100,"value": "","valueProb": 100},{"key": "校验码","keyProb": 100,"value": "","valueProb": 100},{"key": "受票方名称","keyProb": 100,"value": "百望股份有限公司","valuePos": [{"x": 54,"y": 89},{"x": 133,"y": 89},{"x": 133,"y": 101},{"x": 54,"y": 101}],"valueProb": 100},{"key": "受票方税号","keyProb": 100,"value": "91110108339805094M","valuePos": [{"x": 159,"y": 119},{"x": 303,"y": 119},{"x": 303,"y": 135},{"x": 159,"y": 135}],"valueProb": 100},{"key": "受票方地址、电话","keyProb": 100,"value": "","valueProb": 100},{"key": "受票方开户行、账号","keyProb": 100,"value": "","valueProb": 100},{"key": "密码区","keyProb": 100,"value": "","valueProb": 100},{"key": "不含税金额","keyProb": 100,"value": "350.94","valuePos": [{"x": 433,"y": 270},{"x": 469,"y": 270},{"x": 469,"y": 280},{"x": 433,"y": 280}],"valueProb": 100},{"key": "发票税额","keyProb": 95,"value": "21.05","valuePos": [{"x": 598,"y": 270},{"x": 629,"y": 270},{"x": 629,"y": 280},{"x": 598,"y": 280}],"valueProb": 95},{"key": "大写金额","keyProb": 100,"value": "叁佰柒拾贰圆整","valuePos": [{"x": 186,"y": 287},{"x": 257,"y": 287},{"x": 257,"y": 300},{"x": 186,"y": 300}],"valueProb": 100},{"key": "发票金额","keyProb": 100,"value": "372.00","valuePos": [{"x": 472,"y": 287},{"x": 518,"y": 287},{"x": 518,"y": 299},{"x": 472,"y": 299}],"valueProb": 100},{"key": "销售方名称","keyProb": 100,"value": "华住酒店管理有限公司大连分公司","valuePos": [{"x": 362,"y": 88},{"x": 512,"y": 88},{"x": 512,"y": 101},{"x": 362,"y": 101}],"valueProb": 100},{"key": "销售方税号","keyProb": 100,"value": "91210202071550843F","valuePos": [{"x": 470,"y": 119},{"x": 613,"y": 119},{"x": 613,"y": 135},{"x": 470,"y": 135}],"valueProb": 100},{"key": "销售方地址、电话","keyProb": 100,"value": "","valueProb": 100},{"key": "销售方开户行、账号","keyProb": 100,"value": "","valueProb": 100},{"key": "收款人","keyProb": 100,"value": "","valueProb": 100},{"key": "复核人","keyProb": 100,"value": "","valueProb": 100},{"key": "开票人","keyProb": 100,"value": "辛梦娣","valuePos": [{"x": 91,"y": 385},{"x": 122,"y": 385},{"x": 122,"y": 397},{"x": 91,"y": 397}],"valueProb": 100},{"key": "备注","keyProb": 100,"value": "","valueProb": 100},{"key": "标题","keyProb": 100,"value": "电子发票(增值税专用发票)","valuePos": [{"x": 169,"y": 7},{"x": 442,"y": 8},{"x": 442,"y": 34},{"x": 169,"y": 33}],"valueProb": 100},{"key": "联次","keyProb": 100,"value": "","valueProb": 100},{"key": "发票类型","keyProb": 100,"value": "数电专用发票","valuePos": [{"x": 169,"y": 7},{"x": 442,"y": 8},{"x": 442,"y": 34},{"x": 169,"y": 33}],"valueProb": 100},{"key": "特殊标识信息","keyProb": 100,"value": "","valueProb": 100},{"key": "发票详单","keyProb": 100,"value": "[{\"货物或应税劳务、服务名称\":\"*住宿服务*住宿费\",\"规格型号\":\"\",\"单位\":\"天\",\"数量\":\"1\",\"单价\":\"350.943396226415\",\"金额\":\"350.94\",\"税率\":\"6%\",\"税额\":\"21.06\"}]","valueProb": 100},{"key": "发票代码解析","keyProb": 100,"value": "","valueProb": 100}],"sid": "f0626b3869d9156bc8fc3640aba319bfe29db5b59764111e282b986b2d0ac5e9724d7263","width": 660
}

下一步根据将识别结果和表单字段对应，从而返回给前端。

2. Tesseract OCR

Tesseract OCR 是一款开源的文本识别（OCR）引擎。它主要用于识别图片中的文字，并将其转换为可编辑的文本。Tesseract OCR 是目前公认最优秀、最精确的开源 OCR 系统之一。
使用 Spring Boot 框架构建 RESTful API，并集成了 Tesseract OCR 引擎进行文字识别。项目地址：

3. 利用java调用大模型进行识别

基于 LangChain 0.36.0 版本

4. 飞桨PaddleOCR

https://gitee.com/paddlepaddle/PaddleOCR
云服务器购买
linux安装conda环境:

在 Ubuntu 系统中安装 Conda 环境，你可以通过以下几个步骤来完成。Conda 是一个开源的包、依赖和环境管理器，用于安装和管理科学计算相关的软件包和库。它特别适用于 Python 编程语言，但同样可以用于其他语言。步骤 1: 更新你的系统首先，确保你的 Ubuntu 系统是最新的。打开终端并运行以下命令：sudo apt update
sudo apt upgrade
步骤 2: 安装 MinicondaMiniconda 是 Conda 的一个轻量级版本，包含了 conda、Python 和一些常用的包。你可以选择安装 Miniconda 来开始你的 Conda 环境之旅。下载 Miniconda访问 Miniconda 的官方网站（https://docs.conda.io/en/latest/miniconda.html）选择适合你的 Ubuntu 版本的 Miniconda 安装器。例如，对于 64 位系统，你可以使用：wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh运行安装脚本使用以下命令运行下载的安装脚本：bash Miniconda3-latest-Linux-x86_64.sh在运行脚本时，它会提示你阅读许可协议并询问是否接受。按 Enter 键接受许可协议，然后按 yes 确认继续安装。初始化 Conda按照脚本指示，运行以下命令来初始化 Conda：source ~/.bashrc步骤 3: 验证 Conda 安装为了验证 Conda 是否成功安装，你可以在终端中运行以下命令：conda --version这应该会显示已安装的 Conda 版本号。步骤 4: 创建和管理环境（可选）一旦 Conda 安装完成，你可以开始创建和管理环境了。例如，创建一个新的 Python 环境：conda create --name myenv python=3.8激活这个环境：conda activate myenv安装包到特定环境中：conda install numpy
步骤 5: 更新 Conda（可选）为了保持你的 Conda 安装是最新的，定期运行以下命令来更新 Conda：conda update conda通过以上步骤，你应该能够在 Ubuntu 系统上成功安装并使用 Conda 环境。

linux进入/退出conda环境-ubuntu：

使用conda activate base可以切换到conda的基础环境。
使用conda deactivate可以退出当前激活的conda环境，回到系统的默认shell环境。