计算机应用研究2024，Vol.41Issue(11)：3484-3489,6.DOI:10.19734/j.issn.1001-3695.2024.01.0067
TRNet:基于遥感影像的三通道区域增强变化检测网络
TRNet:triple-channel region-enhancement network for change detection based on remote sensing image
摘要
Abstract
Remote sensing image change detection is one of the research focuses in the field of remote sensing.At present,most of them are deep learning methods,which mainly use single channel or siamese network to extract features,which can ef-fectively extract change features.However,with the increasing resolution of remote sensing images,the single feature extrac-tion method is susceptible to the influence of irrelevant details,which leads to the insufficient segmentation ability of the changed and unchanged regions in the detection results.Therefore,this paper proposed a fire-new triple-channel region-enhancement change detection network to enhance feature extraction capability from multiple perspectives.Firstly,the method constructed a triple-channel region-enhancement encoder,and used three feature extraction channels to extract similarity infor-mation,comprehensiveness information and difference information in a directional manner.At every scales of the encoder,region-separation enhancement modules were able to augment channel 2 with channels 1 and 3,which was beneficial to obtain better effect of changing region segmentation.Secondly,it designed a layer interaction-guidance fusion decoder,and used the interactive guidance between higher-level and lower-level features.So the decoder fused effectively the different-level features by the mutual guidance between high-level features and low-level features.Finally,it used an adaptive weight based on infor-mation entropy,which gave more attention to high entropy regions,to optimize loss function.Then,the new loss function im-proved the training process of the network model.The results of experiment on common datasets show that this network achieves better detection accuracy than the contrast networks.关键词
遥感/变化检测/深度学习/区域增强/层级交互Key words
remote sensing/change detection/deep learning/region enhancement/layer interaction分类
信息技术与安全科学引用本文复制引用
石卫超,宋宝贵,管宗胜,秦道龙,邵攀..TRNet:基于遥感影像的三通道区域增强变化检测网络[J].计算机应用研究,2024,41(11):3484-3489,6.基金项目
国家自然科学基金资助项目(41901341,42101469) （41901341,42101469）
湖北省自然科学基金资助项目(2024AFB867)本文受到三峡大学先进计算中心(Advanced Com-putting Center,China Three Gorges University)资助支持. （2024AFB867）
