{"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/deep-faster-detection-of-faint-edges-in-noisy","title":"Multi-scale Processing of Noisy Images using Edge Preservation Losses","arxiv_id":"1803.09420","date":"2018-03-26","proceeding":null,"authors":["Nati Ofir","Yosi Keller"],"abstract":"Noisy images processing is a fundamental task of computer vision. The first\nexample is the detection of faint edges in noisy images, a challenging problem\nstudied in the last decades. A recent study introduced a fast method to detect\nfaint edges in the highest accuracy among all the existing approaches. Their\ncomplexity is nearly linear in the image's pixels and their runtime is seconds\nfor a noisy image. Their approach utilizes a multi-scale binary partitioning of\nthe image. By utilizing the multi-scale U-net architecture, we show in this\npaper that their method can be dramatically improved in both aspects of run\ntime and accuracy. By training the network on a dataset of binary images, we\ndeveloped an approach for faint edge detection that works in a linear\ncomplexity. Our runtime of a noisy image is milliseconds on a GPU. Even though\nour method is orders of magnitude faster, we still achieve higher accuracy of\ndetection under many challenging scenarios. In addition, we show that our\napproach to performing multi-scale preprocessing of noisy images using U-net\nimproves the ability to perform other vision tasks under the presence of noise.\nWe prove it on the problems of noisy objects classification and classical image\ndenoising. We show that multi-scale denoising can be carried out by a novel\nedge preservation loss. As our experiments show, we achieve high-quality\nresults in the three aspects of faint edge detection, noisy image\nclassification and natural image denoising.","url_abs":"http://arxiv.org/abs/1803.09420v5","url_pdf":"http://arxiv.org/pdf/1803.09420v5.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":"deep-faster-detection-of-faint-edges-in-noisy","repo_url":"https://github.com/NatiOfir/DeepFaintEdges","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"edge-detection","task_name":"Edge Detection"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-denoising","task_name":"Image Denoising"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"u-net","method_name":"U-Net"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}