Spectral‐spatial sequence characteristics‐based convolutional transformer for hyperspectral change detection

作     者：Chengle Zhou Qian Shi Da He Bing Tu Haoyang Li Antonio Plaza 

作者机构：School of Geography and PlanningSun Yat‐sen UniversityGuangzhouChina Guangdong Provincial Key Laboratory for Urbanization and GeoSimulationSun Yat‐sen UniversityGuangzhouChina Institute of Optics and ElectronicsNanjing University of Information Science and TechnologyNanjingChina Hyperspectral Computing LaboratoryEscuela PolitecnicaUniversity of ExtremaduraCaceresSpain 

出 版 物：《CAAI Transactions on Intelligence Technology》 (智能技术学报（英文）)

年 卷 期：2023年第8卷第4期

页      面：1237-1257页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：supported in part by by the National Key R&D Program of China under Grant 2022YFB3903402 in part by the National Natural Science Foundation of China under Grant 42222106 in part by the National Natural Science Foundation of China under Grant 61976234 and 42201340 

主　　题：change detection imageanalysis 

摘      要：Recently,ground coverings change detection(CD)driven by bitemporal hyperspectral images(HSIs)has become a hot topic in the remote sensing *** are two challenges in the HSI‐CD task:(1)attribute feature representation of pixel pairs and(2)feature extraction of attribute patterns of pixel *** solve the above problems,a novel spectral‐spatial sequence characteristics‐based convolutional transformer(S3C‐CT)method is proposed for the HSI‐CD *** the designed method,firstly,an eigenvalue extrema‐based band selection strategy is introduced to pick up spectral information with salient attribute ***,a 3D tensor with spectral‐spatial sequence characteristics is proposed to represent the attribute features of pixel pairs in the bitemporal ***,a fusion framework of the convolutional neural network(CNN)and Transformer encoder(TE)is designed to extract high‐order sequence semantic features,taking into account both local context information and global sequence ***,a spatial‐spectral attention mechanism is employed to prevent information reduction and enhance dimensional interactivity between the CNN and ***,the binary change map is determined according to the fully‐connected *** results on real HSI datasets indicated that the proposed S3C‐CT method outperforms other well‐known and state‐of‐the‐art detection approaches in terms of detection performance.