一种屏摄表单图像文本识别方法、装置、设备及存储介质
By combining moiré pattern elimination, image exposure, and table detection models with optical character recognition technology, the moiré pattern problem in low-quality screen-captured form image recognition was solved, improving recognition accuracy and efficiency.
CN117095417BActive Publication Date: 2026-07-17ZHEJIANG YIQICAIYING NETWORK TECHNOLOGY CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG YIQICAIYING NETWORK TECHNOLOGY CO LTD
- Filing Date
- 2023-08-24
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies struggle to effectively eliminate moiré patterns in low-quality screen-captured form images, leading to missed or false detections in OCR model recognition results, resulting in low recognition accuracy and efficiency.
Method used
Moiré patterns are eliminated using a preset moiré pattern elimination model, image exposure is processed and perspective transformation is performed, cell vertex coordinates are determined using a table detection model, and text detection is performed on the form area using optical character recognition technology.
Benefits of technology
It improves the accuracy and efficiency of text recognition in screen-captured form images, reduces false positives and false negatives, and enhances the accuracy and speed of recognition.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117095417B_ABST
Abstract
本申请公开了一种屏摄表单图像文本识别方法、装置、设备及存储介质，涉及图像识别技术领域，包括：将目标屏摄表单图像输入至预设摩尔纹消除模型以得到去掉摩尔纹的第一目标像素矩阵，并对所述第一目标像素矩阵进行曝光处理以确定与所述目标屏摄表单图像上的目标表单区域对应的第二目标像素矩阵；将所述第二目标像素矩阵输入至预设表格检测模型以得到目标单元格顶点坐标，并基于所述目标单元格顶点坐标确定单元格像素矩阵；对所述单元格像素矩阵中的单元格进行拼接得到目标图像，利用预设光学字符识别技术对所述目标图像进行文本检测得到所述目标表单区域上的文本。这样一来，实现了自动化的图像校准和文字识别，减少了人工干预，提高了识别效率。
Need to check novelty before this filing date? Find Prior Art