Table structure recognition based on dual-branch encoder-decoder segmentation
Uložené v:
| Názov: | Table structure recognition based on dual-branch encoder-decoder segmentation |
|---|---|
| Autori: | Dajun Xiao, Xialing Xu, Yue Zhang, Tao Liu, Xin Li, Yongtian Qiao |
| Zdroj: | Journal of Computational Methods in Sciences and Engineering. |
| Informácie o vydavateľovi: | SAGE Publications, 2025. |
| Rok vydania: | 2025 |
| Popis: | Digitization of paper documents is crucial for the management of modern power grid enterprise. Table structure recognition, which identifies table cells, presents challenges due to diverse table formats. This paper introduces a novel table structure recognition method based on dual-branch encoder-decoder segmentation. The proposed approach converts table structure extraction into row and column segmentation sub-problems, which utilizes a single encoder for feature extraction and two independent decoder branches for segment prediction. In this framework, a Conv-Res-CBAM unit is proposed to enhance feature extraction and transmission. Additionally, the Tesseract OCR engine is incorporated for character recognition. Extensive experiments on two public datasets and a self-collected dataset demonstrate the superiority of our method. |
| Druh dokumentu: | Article |
| Jazyk: | English |
| ISSN: | 1875-8983 1472-7978 |
| DOI: | 10.1177/14727978251361868 |
| Rights: | URL: https://journals.sagepub.com/page/policies/text-and-data-mining-license |
| Prístupové číslo: | edsair.doi...........b1227fea5aad47095f86f532e672d91d |
| Databáza: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstrakt: | Digitization of paper documents is crucial for the management of modern power grid enterprise. Table structure recognition, which identifies table cells, presents challenges due to diverse table formats. This paper introduces a novel table structure recognition method based on dual-branch encoder-decoder segmentation. The proposed approach converts table structure extraction into row and column segmentation sub-problems, which utilizes a single encoder for feature extraction and two independent decoder branches for segment prediction. In this framework, a Conv-Res-CBAM unit is proposed to enhance feature extraction and transmission. Additionally, the Tesseract OCR engine is incorporated for character recognition. Extensive experiments on two public datasets and a self-collected dataset demonstrate the superiority of our method. |
|---|---|
| ISSN: | 18758983 14727978 |
| DOI: | 10.1177/14727978251361868 |