{"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/pixel-level-reconstruction-and-classification","title":"Pixel-level Reconstruction and Classification for Noisy Handwritten Bangla Characters","arxiv_id":"1806.08037","date":"2018-06-21","proceeding":null,"authors":["Manohar Karki","Qun Liu","Robert DiBiano","Saikat Basu","Supratik Mukhopadhyay"],"abstract":"Classification techniques for images of handwritten characters are\nsusceptible to noise. Quadtrees can be an efficient representation for learning\nfrom sparse features. In this paper, we improve the effectiveness of\nprobabilistic quadtrees by using a pixel level classifier to extract the\ncharacter pixels and remove noise from handwritten character images. The pixel\nlevel denoiser (a deep belief network) uses the map responses obtained from a\npretrained CNN as features for reconstructing the characters eliminating noise.\nWe experimentally demonstrate the effectiveness of our approach by\nreconstructing and classifying a noisy version of handwritten Bangla Numeral\nand Basic Character datasets.","url_abs":"http://arxiv.org/abs/1806.08037v1","url_pdf":"http://arxiv.org/pdf/1806.08037v1.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":[],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"document-image-classification","task_name":"Document Image Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/document-image-classification-on-noisy-bangla-1","task":"Document Image Classification","dataset":"Noisy Bangla Characters","model":"Pixel-level RC","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"77.22"},"uses_additional_data":false},{"leaderboard":"/sota/document-image-classification-on-noisy-bangla","task":"Document Image Classification","dataset":"Noisy Bangla Numeral","model":"Pixel-level RC","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"95.46"},"uses_additional_data":false},{"leaderboard":"/sota/document-image-classification-on-noisy-mnist","task":"Document Image Classification","dataset":"n-MNIST","model":"Pixel-level RC","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"97.62"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-noisy-mnist-awgn","task":"Image Classification","dataset":"Noisy MNIST (AWGN)","model":"Pixel-level RC","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"97.62"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-noisy-mnist-contrast","task":"Image Classification","dataset":"Noisy MNIST (Contrast)","model":"Pixel-level RC","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"95.04"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-noisy-mnist-motion","task":"Image Classification","dataset":"Noisy MNIST (Motion)","model":"Pixel-level RC","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"97.20"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}