{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/binary-document-image-super-resolution-for","title":"Binary Document Image Super Resolution for Improved Readability and OCR Performance","arxiv_id":"1812.02475","date":"2018-12-06","proceeding":null,"authors":["Ram Krishna Pandey","K Vignesh","A. G. Ramakrishnan","Chandrahasa B"],"abstract":"There is a need for information retrieval from large collections of\nlow-resolution (LR) binary document images, which can be found in digital\nlibraries across the world, where the high-resolution (HR) counterpart is not\navailable. This gives rise to the problem of binary document image\nsuper-resolution (BDISR). The objective of this paper is to address the\ninteresting and challenging problem of super resolution of binary Tamil\ndocument images for improved readability and better optical character\nrecognition (OCR). We propose multiple deep neural network architectures to\naddress this problem and analyze their performance. The proposed models are all\nsingle image super-resolution techniques, which learn a generalized spatial\ncorrespondence between the LR and HR binary document images. We employ\nconvolutional layers for feature extraction followed by transposed convolution\nand sub-pixel convolution layers for upscaling the features. Since the outputs\nof the neural networks are gray scale, we utilize the advantage of power law\ntransformation as a post-processing technique to improve the character level\npixel connectivity. The performance of our models is evaluated by comparing the\nOCR accuracies and the mean opinion scores given by human evaluators on LR\nimages and the corresponding model-generated HR images.","url_abs":"http://arxiv.org/abs/1812.02475v1","url_pdf":"http://arxiv.org/pdf/1812.02475v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"binary-document-image-super-resolution-for","repo_url":"https://github.com/gregbugaj/unet-denoiser","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mxnet","reach":null}],"tasks":[{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"optical-character-recognition","task_name":"Optical Character Recognition"},{"task_slug":"optical-character-recognition","task_name":"Optical Character Recognition (OCR)"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.02475","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}