- 概述
- 入门指南
- Activities (活动)
- Insights 仪表板
- Document Understanding 流程
- 快速入门教程
- 从收据中提取数据
- 使用附加字段重新训练发票
- 从表单中提取数据
- 从文件开始创建新自动化
- 框架组件
- 模型详细信息
- 概述
- Document Understanding - ML 包
- DocumentClassifier - ML 包
- 具有 OCR 功能的 ML 包
- 1040 - ML 包
- 1040 附表 C - ML 包
- 1040 附表 D - ML 包
- 1040 附表 E - ML 包
- 1040x - ML 包
- 3949a - ML 包
- 4506T - ML 包
- 709 - ML 包
- 941x - ML 包
- 9465 - ML 包
- ACORD125 - ML 包
- ACORD126 - ML 包
- ACORD131 - ML 包
- ACORD140 - ML 包
- ACORD25 - ML 包
- 银行对账单 - ML 包
- 提单 - ML 包
- 公司注册证书 - ML 包
- 原产地证书 - ML 包
- 检查 - ML 包
- 儿童产品证书 - ML 包
- CMS1500 - ML 包
- 欧盟符合性声明 - ML 包
- 财务报表 (Financial statements) - ML 包
- FM1003 - ML 包
- I9 - ML 包
- ID Cards - ML 包
- Invoices - ML 包
- InvoicesAustralia - ML 包
- 中国发票 - ML 包
- 希伯来语发票 - ML 包
- 印度发票 - ML 包
- 日本发票 - ML 包
- 装运发票 - ML 包
- 装箱单 - ML 包
- 工资单 - ML 包
- 护照 - ML 包
- 采购订单 - ML 包
- 收据 - ML 包
- 汇款通知书 - ML 包
- UB04 - ML 包
- 水电费账单 - ML 包
- 车辆所有权证明 - ML 包
- W2 - ML 包
- W9 - ML 包
- 其他开箱即用的 ML 包
- 公共端点
- 流量限制
- OCR 配置
- 管道
- OCR 服务
- 支持的语言
- 深度学习
- 数据与安全性
- 许可和计费逻辑

Document Understanding classic user guide
从收据中提取数据
The aim of this page is to help first time users get familiar with UiPath® Document UnderstandingTM. For scalable production deployments, we strongly recommend using the Document Understanding Process available in UiPath® Studio under the Templates section.
This quickstart shows you how to extract data from receipts using the out-of-the-box Receipts ML model with its corresponding public endpoint.
可以通过提供“验证站点”或使用 Action Center 中的“验证操作”来完成验证。以下各部分将介绍这两个选项。
将收据 ML 模型与公共端点和验证站点一起使用
在本节中,我们将使用“验证站点”来验证提取结果。
To create a basic workflow using the Receipts ML Model, use the following steps:
- 创建空白流程
- 安装所需的活动包
- 创建分类
- 将文档数字化
- 使用收据 ML 模型提取数据
- 使用验证站点验证结果
- 导出提取结果
1. 创建空白流程
- 启动 UiPath Studio。
- In the HOME backstage view, select Process to create a new project.
- 系统将显示“新的空白流程”窗口。在此窗口中,输入新项目的名称。如果需要,您还可以添加说明,以便更轻松地对项目进行排序。
- Select Create. The new project is opened in Studio.
2. 安装所需的活动包
From the Manage Packages button in the ribbon, besides the core activities packages (UiPath.Excel.Activities, UiPath.Mail.Activities, UiPath.System.Activities, UiPath.UIAutomation.Activities) that are added to the project by default, install the following activities packages:
- UiPath.DocumentUnderstanding.ML.Activities
- UiPath.DocumentUnderstanding.OCR.LocalServer
- UiPath.Intelligent OCR.Activities
- UiPath.MLServices.Activities
3. 创建分类
Once the activities packages are installed, list out the required fields. The Receipts ML model supports data extraction for the following fields:
- 名称 -
Text - 供应商地址 -
Address - 总计 -
Number - 日期 -
Date - 电话号码 -
Text - 货币 -
Text - 费用类型 -
Text - 项目 -
Table- 说明 -
Text - 行数量 -
Number - 单价 -
Number - 数量 -
Number
- 说明 -
Open Taxonomy Manager and create a group named Semi Structured Documents, a category named Finance, and a document type named Receipts. Create the listed fields with user friendly names along with respective data types.

4. 将文档数字化
- In the Main.xaml file, add a Load Taxonomy activity and create a variable for the taxonomy output.
- Add a Digitize Document activity with UiPath Document OCR. Provide the input property Document Path and create output variables for Document Text and Document Object Model.
- 请记住在“UiPath 文档 OCR”活动中添加 Document Understanding API 密钥。
5. 使用收据 ML 模型提取数据
-
Add a Data Extraction Scope activity and fill in the properties.
-
Drag and drop a Machine Learning Extractor activity. A pop-up with three input parameters, Endpoint, ML Skill, and ApiKey, is displayed on the screen.
-
Fill in the Endpoint parameter with the Receipts Public Endpoint, namely
https://du.uipath.com/ie/receipts, and provide the Document Understanding API key. -
Select Get Capabilities.
-
The next step is to configure the extractor. Configuring the extractor means mapping the fields that you created in Taxonomy Manager to the fields available in the ML model.

-
要将机器学习提取程序与 ML 技能一起使用,请从下拉列表中选择 ML 技能并配置提取程序。
-
您必须将机器人 Assistant 连接到与 ML 技能所在的同一租户。

6. 使用验证站点验证结果
To check the results through Validation Station, drag and drop the Present Validation Station activity and provide the input details.

7. 导出提取结果
To export the extraction results, drag and drop an Export Extraction Results activity to the end of your workflow. This outputs the results into a DataSet that contains multiple tables, which could then be written to an Excel file or be used directly in a downstream process.

下载示例
Download this sample project using this link.
该示例包含两个工作流:
- Main.xaml - in this workflow, the extraction results are validated using Validation Station
- Main - Unattended.xaml - 在此工作流中,使用“验证操作”验证提取结果;下一节将对此进行说明
将收据 ML 模型与公共端点和验证操作一起使用
请参阅此页面,详细了解如何使用 Action Center 验证操作,而非显示验证站点。
Action Center 中的任务如何工作?
当自动化包括应人工做出的决策(例如批准、升级和异常处理)时,UiPath Action Center 可以轻松高效地将流程从机器人移交给人工。然后再次返回。
Document Understanding Action Center 活动随附 UiPath.IntelligentOCR.Activities 包和 UiPath.Persistance.Activities 包。请勿忘记从 UiPath Studio 的“常规设置”中启用“暂留”活动:

验证操作如何工作?
通过在内部部署 Orchestrator 和 Automation Cloud 中创建可在 Action Center 中添加文档验证操作的编排流程来提高生产力。此操作可减少在本地存储文档、在每个人类操作的计算机上安装一个机器人或让机器人等待人类用户完成验证的需求。
More details here.
如何使用验证操作?
重复上一部分中所述的第 1 步至第 5 步。
Then, instead of using the Present Validation Station activity, use the Create Document Validation Action and Wait for Document Validation Action and Resume activities.

This creates a document validation action in Action Center. The output of the Create Document Validation Action activity can then be used with the Wait for Document Validation Action and Resume activity to suspend and resume orchestration workflows upon human action completion in Action Center.