{"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/going-deeper-with-contextual-cnn-for","title":"Going Deeper with Contextual CNN for Hyperspectral Image Classification","arxiv_id":"1604.03519","date":"2016-04-12","proceeding":null,"authors":["Hyungtae Lee","Heesung Kwon"],"abstract":"In this paper, we describe a novel deep convolutional neural network (CNN)\nthat is deeper and wider than other existing deep networks for hyperspectral\nimage classification. Unlike current state-of-the-art approaches in CNN-based\nhyperspectral image classification, the proposed network, called contextual\ndeep CNN, can optimally explore local contextual interactions by jointly\nexploiting local spatio-spectral relationships of neighboring individual pixel\nvectors. The joint exploitation of the spatio-spectral information is achieved\nby a multi-scale convolutional filter bank used as an initial component of the\nproposed CNN pipeline. The initial spatial and spectral feature maps obtained\nfrom the multi-scale filter bank are then combined together to form a joint\nspatio-spectral feature map. The joint feature map representing rich spectral\nand spatial properties of the hyperspectral image is then fed through a fully\nconvolutional network that eventually predicts the corresponding label of each\npixel vector. The proposed approach is tested on three benchmark datasets: the\nIndian Pines dataset, the Salinas dataset and the University of Pavia dataset.\nPerformance comparison shows enhanced classification performance of the\nproposed approach over the current state-of-the-art on the three datasets.","url_abs":"http://arxiv.org/abs/1604.03519v3","url_pdf":"http://arxiv.org/pdf/1604.03519v3.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":"going-deeper-with-contextual-cnn-for","repo_url":"https://github.com/eecn/Hyperspectral-Classification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"going-deeper-with-contextual-cnn-for","repo_url":"https://github.com/nshaud/DeepHyperX","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"hyperspectral-image-classification","task_name":"Hyperspectral Image Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1604.03519","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}