Semantic segmentation for remote sensing images based on an AD-HRNet model
作者机构：School of Geography and Information EngineeringChina University of GeosciencesWuhanPeople’s Republic of China Wuhan Geomatics InstituteWuhanPeople’s Republic of China State Key Laboratory of Information Engineering in SurveyingMappingand Remote SensingWuhan UniversityWuhanPeople’s Republic of China
出 版 物：《International Journal of Digital Earth》 (国际数字地球学报（英文）)
年 卷 期：2022年第15卷第1期
页 面：2376-2399页
核心收录：
基 金：supported by the National Natural Science Foundation of China(No.42271449,41901394,41971405) open research fund program of Geomatics Technology and Application Key Laboratory of Qinghai Province
主 题：Semantic segmentation convolutional neural networks dilated convolution attention mechanism remote sensing
摘 要：Semantic segmentation for remote sensing images faces challenges of unbalanced category weight,rich context causing difficulties of recognition,blurred boundaries of multi-scale objects,and so *** address these problems,we propose a new model by combining HRNet with attention mechanisms and dilated convolution,denoted as:AD-HRNet for the semantic segmentation of remote sensing *** the framework of AD-HRNet,we obtained the weight value of each category based on an improved weighted cross-entropy function by introducing the median frequency balance method to solve the issue of class weight *** Shuffle-CBAM module with channel attention and spatial attention in AD-HRNet framework was applied to extract more global context information of images through slightly increasing the amount of *** address the problem of blurred boundaries caused by multi-scale object segmentation and edge segmentation,we developed an MDC-DUC module in AD-HRNet framework to capture the context information of multi-scale objects and the edge information of many irregular *** Postdam,Vaihingen,and SAMA-VTOL datasets as materials,we verified the performance of AD-HRNet by comparing with eight typical semantic segmentation *** results shown that AD-HRNet increases the mIoUs to 75.59%and 71.58%based on the Postdam and Vaihingen datasets,respectively.
维普期刊数据库