一种基于OCR和NER模型的影像附件审核方法、设备及介质

By combining OCR and NER models, the text and layout features of image attachments are identified, feature extraction templates are loaded, and a rule engine is used for automatic review. This solves the problem of unstructured image attachments being difficult to parse automatically, and achieves efficient automation and accuracy in financial review.

CN120954037BActive Publication Date: 2026-07-17INSPUR GENERSOFT CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
INSPUR GENERSOFT CO LTD
Filing Date
2025-07-30
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

The existing financial audit system cannot effectively and automatically parse unstructured image attachments, resulting in low approval efficiency and a high risk of errors.

Method used

By combining OCR and NER models, the textual structure and layout features of image attachments are identified, a pre-set feature extraction template is loaded, and a rule engine is used for automatic review to generate a review report.

Benefits of technology

It has enabled automated review of multi-format image attachments, reduced manual intervention, improved review efficiency and reduced error rate, and achieved full automation of the financial review process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120954037B_ABST
    Figure CN120954037B_ABST
Patent Text Reader

Abstract

本申请公开了一种基于OCR和NER模型的影像附件审核方法、设备及介质，方法包括：确定用户上传的影像附件对应的文件类型；根据各文件类型对应的识别模式，对影像附件进行影像识别，以提取影像附件中的文本结构化特征和版式特征；对文本结构化特征进行预处理，并将预处理后的文本结构化特征与版式特征进行融合，得到融合特征；加载预置的要素抽取模板，基于NER模型，从融合特征中抽取与要素抽取模板相匹配的审核要素；通过预置的规则引擎，对审核要素进行规则校验，以生成影像附件对应的审核报告，并将审核报告反馈至用户。
Need to check novelty before this filing date? Find Prior Art