{"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-underwater-image-enhancement","title":"Deep Underwater Image Enhancement","arxiv_id":"1807.03528","date":"2018-07-10","proceeding":null,"authors":["Saeed Anwar","Chongyi Li","Fatih Porikli"],"abstract":"In an underwater scene, wavelength-dependent light absorption and scattering\ndegrade the visibility of images, causing low contrast and distorted color\ncasts. To address this problem, we propose a convolutional neural network based\nimage enhancement model, i.e., UWCNN, which is trained efficiently using a\nsynthetic underwater image database. Unlike the existing works that require the\nparameters of underwater imaging model estimation or impose inflexible\nframeworks applicable only for specific scenes, our model directly reconstructs\nthe clear latent underwater image by leveraging on an automatic end-to-end and\ndata-driven training mechanism. Compliant with underwater imaging models and\noptical properties of underwater scenes, we first synthesize ten different\nmarine image databases. Then, we separately train multiple UWCNN models for\neach underwater image formation type. Experimental results on real-world and\nsynthetic underwater images demonstrate that the presented method generalizes\nwell on different underwater scenes and outperforms the existing methods both\nqualitatively and quantitatively. Besides, we conduct an ablation study to\ndemonstrate the effect of each component in our network.","url_abs":"http://arxiv.org/abs/1807.03528v1","url_pdf":"http://arxiv.org/pdf/1807.03528v1.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-underwater-image-enhancement","repo_url":"https://github.com/pritishuplavikar/All-In-One-Underwater-Image-Enhancement-using-Domain-Adversarial-Learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-underwater-image-enhancement","repo_url":"https://github.com/saeed-anwar/UWCNN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-enhancement","task_name":"Image Enhancement"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}