SegDetector: A Deep Learning Model for Detecting Small and Overlapping Damaged Buildings in Satellite Images

• Published in: Remote sensing (Basel, Switzerland) Vol. 14; no. 23; p. 6136
• Main Authors: Yu, Zhengbo, Chen, Zhe, Sun, Zhongchang, Guo, Huadong, Leng, Bo, He, Ziqiong, Yang, Jinpei, Xing, Shuwen
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.12.2022
• Subjects: Algorithms; Buildings; Classification; Damage assessment; Damage detection; damaged building; Datasets; Deep learning; Disaster relief; Disasters; Effectiveness; Image segmentation; Localization; Machine learning; Natural disasters; overlapping target; Remote sensing; Satellite imagery; Satellites; SegDetector; Semantics; small target; Target detection; Volcanoes

Buildings bear much of the damage from natural disasters, and determining the extent of this damage is of great importance to post-disaster emergency relief. The application of deep learning to satellite remote sensing imagery has become more and more mature in monitoring natural disasters, but ther...