{"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-neural-network-based-sparse-measurement","title":"Deep neural network based sparse measurement matrix for image compressed sensing","arxiv_id":"1806.07026","date":"2018-06-19","proceeding":null,"authors":["Wenxue Cui","Feng Jiang","Xinwei Gao","Wen Tao","Debin Zhao"],"abstract":"Gaussian random matrix (GRM) has been widely used to generate linear\nmeasurements in compressed sensing (CS) of natural images. However, there\nactually exist two disadvantages with GRM in practice. One is that GRM has\nlarge memory requirement and high computational complexity, which restrict the\napplications of CS. Another is that the CS measurements randomly obtained by\nGRM cannot provide sufficient reconstruction performances. In this paper, a\nDeep neural network based Sparse Measurement Matrix (DSMM) is learned by the\nproposed convolutional network to reduce the sampling computational complexity\nand improve the CS reconstruction performance. Two sub networks are included in\nthe proposed network, which are the sampling sub-network and the reconstruction\nsub-network. In the sampling sub-network, the sparsity and the normalization\nare both considered by the limitation of the storage and the computational\ncomplexity. In order to improve the CS reconstruction performance, a\nreconstruction sub-network are introduced to help enhance the sampling\nsub-network. So by the offline iterative training of the proposed end-to-end\nnetwork, the DSMM is generated for accurate measurement and excellent\nreconstruction. Experimental results demonstrate that the proposed DSMM\noutperforms GRM greatly on representative CS reconstruction methods","url_abs":"http://arxiv.org/abs/1806.07026v1","url_pdf":"http://arxiv.org/pdf/1806.07026v1.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-neural-network-based-sparse-measurement","repo_url":"https://github.com/WenxueCui/DSMM","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"image-compressed-sensing","task_name":"Image Compressed Sensing"},{"task_slug":"compressed-sensing","task_name":"compressed sensing"}],"methods":[],"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}