增值税普通发票（卷票）内容识别

对含有单张或多张增值税普通发票（卷票）的图片，提取发票中的发票代码、发票号码、销售方名称、纳税人识别号(销售方)、开票日期、购买方名称、纳税人识别号（购买方）、项目、金额字段所在区域进行检测和识别，获取字段内容。

This dataset is used to process images containing one or more ordinary value-added tax (VAT) roll invoices, by detecting and recognizing the regions where key fields including invoice code, invoice number, seller's name, seller's taxpayer identification number, invoice issuing date, purchaser's name, purchaser's taxpayer identification number, item, and amount field are located in the invoices, and then extracting the content of these fields.

搜集汇总

背景与挑战

背景概述

该数据集专注于增值税普通发票（卷票）的内容识别，能够从包含单张或多张发票的图片中自动检测和提取关键字段，如发票代码、号码、销售方和购买方信息、开票日期、项目及金额。它属于信息传输、软件和信息技术服务业，以远程服务方式提供，旨在实现发票信息的自动化处理，提升数据提取效率。