Enhancing optical character recognition: Efficient techniques for document layout analysis and text line detection

Self Cite

As deep learning continues to evolve, we have observed huge breakthroughs in the fields of medical imaging, video and frame generation, optical character recognition (OCR), and other domains. In the field of data analysis and document processing, the recognition of handwritten numerals plays a crucial role. This work has led to remarkable changes in OCR, historical handwritten document analysis, and postal automation. In this study, we present a novel framework to overcome this challenge, going beyond digit recognition in only one language. Unlike common methods that focus on a limited set of languages, our method provides a comprehensive solution for recognition of handwritten digit images in 12 different languages. These specific languages are chosen because most of them have fairly distant representations in latent space. We utilize transfer learning, as it reduces the computational cost and maintains the quality of enhanced images and the models' recognition accuracy. Another strength of our approach is the innovative attention-based module called the MRA module. Our experiments confirm that by applying this module, major progress is made in both image quality and the accuracy of handwritten digit recognition. Notably, we reached high precisions, surpassing nearly 2% improvement in specific languages compared to earlier techniques. In this work, we present a robust and cost-effective approach that handles multilingual handwritten numeral recognition across a wide range of languages. The code and further implementation details are available at https://github.com/CVLab-SHUT/HandWrittenDigitRecognition

Advancing Multilingual Handwritten Numeral Recognition With Attention-Driven Transfer Learning

Fateh,

Birgani,

Fateh

et al. 2024

Self Cite

Efficient Text Bounding Box Identification Using Mask R-CNN: Case of Thai Documents

Kiatphaisansophon,

Wanvarie,

Cooharojananone

2024

Prototype Comparison Convolutional Networks for One-Shot Segmentation

Li,

Guo

et al. 2024

In few-shot semantic segmentation (FSS), the key challenges are efficiently tuning the interaction between the support set and the query set and distinguishing between context, background, and interfering items. To address these challenges, we propose prototype comparison networks for oneshot segmentation (OPCN) to capture the details required for FSS. Specifically, we offer the Fusion Interaction Module (FIM) to improve the segmentation performance by capturing the correlation and semantic information between the support set and query set features. Subsequently, we propose the Feature Enhancement Module (FEM), which aims to enhance the information required in the support set and query set features while increasing the focus on critical details by reducing the weight of the background regions to provide a more targeted feature representation for subsequent query image segmentation. Then, we propose the Feature Refinement Module (FRM) to filter irrelevant background information in the features and specify the target location region. Finally, the Feature Matching Module (FM) generates the final segmentation mask for the query image. Extensive experiments on the PASCAL-5i and COCO-20i datasets show that our approach achieves excellent performance in the case of the one-shot setup.