{"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/dkdfn-domain-knowledge-guided-deep","title":"DKDFN: Domain Knowledge-Guided deep collaborative fusion network for multimodal unitemporal remote sensing land cover classification","arxiv_id":null,"date":"2022-02-13","proceeding":"ScienseDirect 2022 2","authors":["Liheng Zhong"],"abstract":"Land use and land cover maps provide fundamental information that has been used in different types of studies,\r\nranging from public health to carbon cycling. However, the existing remote sensing image classification methods\r\nthus far suffer from the insufficient usage of multiple modalities, underconsideration of prior domain knowledge,\r\nand poor performance on minority classes. To alleviate these problems, we propose a novel domain knowledge-\r\nguided deep collaborative fusion network (DKDFN) with performance boosting for minority categories for land\r\ncover classification. More specifically, the DKDFN adopts a multihead encoder and a multibranch decoder\r\nstructure. The architecture of the encoder probablizes sufficient mining of complementary information from\r\nmultiple modalities, which are Sentinel-2, Sentinel-1, and SRTM Digital Elevation Data (SRTM) in our case. The\r\nmultibranch decoder enables land cover classification in a multitask learning setup, performing semantic seg­\r\nmentation and reconstructing multimodal remote sensing indices, which are selected as representatives of\r\ndomain knowledge. This design incorporates domain knowledge in an effective end-to-end manner. The training\r\nstage of our DKDFN is supervised by our proposed asymmetry loss function (ALF), which boosts performance on\r\nnearly all categories, especially the categories with a low frequency of occurrence. Ablation studies of the\r\nnetwork suggest that our design logic is worth testing in any network with an encoder-decoder structure. The\r\nstudy is conducted in Hunan, China and is verified using a self-labeled multimodal unitemporal remote sensing\r\nimage dataset. The comparative experiments between DKDFN and 6 state-of-the-art models (U-Net, SegNet,\r\nPSPNet, DeepLab, HRNet, MP-ResNet) testify to the superiority of our method and suggest its potential to be\r\napplied more widely to map land cover in other geographical areas given the availability of Sentinel-2, Sentinel-\r\n1, and SRTM data. The dataset can be downloaded by https://github.com/LauraChow/HunanMultimodalDataset","url_abs":"https://www.sciencedirect.com/science/article/pii/S0924271622000557","url_pdf":"https://doi.org/10.1016/j.isprsjprs.2022.02.013","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":"dkdfn-domain-knowledge-guided-deep","repo_url":"https://github.com/LauraChow/HunanMultimodalDataset","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"land-cover-classification","task_name":"Land Cover Classification"},{"task_slug":"remote-sensing-image-classification","task_name":"Remote Sensing Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}