Pre-training transformer with dual-branch context content module for table detection in document images

作     者：Yongzhi LI Pengle ZHANG Meng SUN Jin HUANG Ruhan HE 

作者机构：School of Computer Science and Artificial Intelligence Wuhan Textile University Wuhan China 430064 School of Computer Science South-Central Minzu University Wuhan China 430064 Hubei Provincial Engineering Research Center for Intelligent Textile and Fashion Wuhan Textile University Wuhan China 430064 

出 版 物：《虚拟现实与智能硬件》 (Virtual Reality & Intelligent Hardware)

年 卷 期：2024年第6卷第5期

页      面：408-420页

核心收录：

学科分类：0401[教育学-教育学] 04[教育学] 

主　　题：Table detection Document image analysis Transformer Dilated convolution Deformable convolution Feature fusion 

摘      要：BackgroundDocument images such as statistical reports and scientific journals are widely used in information technology. Accurate detection of table areas in document images is an essential prerequisite for tasks such as information extraction. However, because of the diversity in the shapes and sizes of tables, existing table detection methods adapted from general object detection algorithms, have not yet achieved satisfactory results. Incorrect detection results might lead to the loss of critical information. MethodsTherefore, we propose a novel end-to-end trainable deep network combined with a self-supervised pretraining transformer for feature extraction to minimize incorrect detections. To better deal with table areas of different shapes and sizes, we added a dual-branch context content attention module (DCCAM) to high-dimensional features to extract context content information, thereby enhancing the network s ability to learn shape features. For feature fusion at different scales, we replaced the original 3×3 convolution with a multilayer residual module, which contains enhanced gradient flow information to improve the feature representation and extraction capability. ResultsWe evaluated our method on public document datasets and compared it with previous methods, which achieved state-of-the-art results in terms of evaluation metrics such as recall and F1-score. https://***/YongZ-Lee/TD-DCCAM