Semantic segmentation for remote sensing images based on an AD-HRNet model  

作　　者：Xue Yang Xiang Fan Mingjun Peng Qingfeng Guan Luliang Tang 

机构地区：[1]School of Geography and Information Engineering,China University of Geosciences,Wuhan,People’s Republic of China [2]Wuhan Geomatics Institute,Wuhan,People’s Republic of China [3]State Key Laboratory of Information Engineering in Surveying,Mapping,and Remote Sensing,Wuhan University,Wuhan,People’s Republic of China

出　　处：《International Journal of Digital Earth》2022年第1期2376-2399,共24页国际数字地球学报（英文）

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.

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 on.To 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 images.In 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 imbalance.The 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 computation.To 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 objects.Taking Postdam,Vaihingen,and SAMA-VTOL datasets as materials,we verified the performance of AD-HRNet by comparing with eight typical semantic segmentation models.Experimental results shown that AD-HRNet increases the mIoUs to 75.59%and 71.58%based on the Postdam and Vaihingen datasets,respectively.

关 键 词：Semantic segmentation convolutional neural networks dilated convolution attention mechanism remote sensing 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]