2026-04-20T04:05:00Z https://ipsj.ixsq.nii.ac.jp/oai

oai:ipsj.ixsq.nii.ac.jp:00216932 2025-01-19T15:40:58Z 1164:4619:10826:10881

Deep Cascade Road Extraction Network:a Multi-task Method for Road Extraction Deep Cascade Road Extraction Network:a Multi-task Method for Road Extraction Yubo, Wang Zhao, Wang Yuusuke, Nakano Ken, Nishimatsu Katsuya, Hasegawa Jun, Ohya Yubo, Wang Zhao, Wang Yuusuke, Nakano Ken, Nishimatsu Katsuya, Hasegawa Jun, Ohya セッション1-A In this work, we present an end-to-end cascade neural network model called Deep Cascade Road Extraction Network to extract accurate road networks from aerial imagery. On the basis of the cascade structure consisting of three subnetworks (Surface Segmentation Network, Edge Detection Network, and Centreline Extraction Network) connected in a cascade manner, we simultaneously achieve the three tasks of the subnetworks for road extraction. Through comparison experiments, our method achieves state-of-the-art results for all the three subtasks. Meanwhile, our model demonstrates strong robustness to occlusions while accurately extracting complex road areas. In this work, we present an end-to-end cascade neural network model called Deep Cascade Road Extraction Network to extract accurate road networks from aerial imagery. On the basis of the cascade structure consisting of three subnetworks (Surface Segmentation Network, Edge Detection Network, and Centreline Extraction Network) connected in a cascade manner, we simultaneously achieve the three tasks of the subnetworks for road extraction. Through comparison experiments, our method achieves state-of-the-art results for all the three subtasks. Meanwhile, our model demonstrates strong robustness to occlusions while accurately extracting complex road areas. 情報処理学会 2022-03-03 eng technical report https://ipsj.ixsq.nii.ac.jp/records/216932 2188-8701 AA11131797 研究報告コンピュータビジョンとイメージメディア（CVIM） 2022-CVIM-229 1 1 6 https://ipsj.ixsq.nii.ac.jp/record/216932/files/IPSJ-CVIM22229001.pdf application/pdf 1.2 MB