Robust Land Cover Classification With Local–Global Information Decoupling to Address Remote Sensing Anomalous Data

被引:3
作者
Xiao, Jianbo [1 ]
Cheng, Taotao [1 ]
Chen, Deliang [1 ]
Chen, Hui [2 ]
Li, Ning [2 ]
Lu, Yanyan [3 ]
Cheng, Liang [2 ]
机构
[1] Nanjing Univ Posts & Telecommun, Sch Geog & Bioinformat, Nanjing 210023, Peoples R China
[2] Nanjing Univ, Sch Geog & Ocean Sci, Nanjing 210023, Peoples R China
[3] Nanjing Audit Univ, Inst Nat Resources & Environm Audit, Nanjing 211815, Peoples R China
关键词
Feature extraction; Remote sensing; Deep learning; Training; Computational modeling; Task analysis; Sensors; land cover classification; remote sensing anomalous data; remote sensing imagery; SEGMENTATION; VEGETATION;
D O I
10.1109/JSTARS.2024.3360458
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Remote sensing images play a critical role in urban planning, land resources, and environmental monitoring. Land cover classification is one of the straightforward applications of remote sensing. However, the anomalous remote sensing data challenges the reliability of land cover classification results. Deep learning has been widely used in remote sensing image analysis, but it remains sensitive to anomalous data. To address this issue, we reevaluate a land cover classification map in high-noise scenarios with anomalous data and propose a novel network architecture to solve the problem. A new network architecture is proposed to solve this problem. Our proposed network architecture focuses on decoupling the extraction of global information and local information. Through three global-local feature fusion modules, we output features emphasizing global information, features emphasizing local information, and consistency evaluation scores, respectively. A specially designed decoder integrates these three features. Our method performs better compared to mainstream models on the public datasets the Wuhan high-definition landscape dataset with obvious anomaly data, with a mean intersection over union (MIoU) of 63.58% and a mean pixel accuracy (Mpa) of 74.32%. Compared to the suboptimal method, our method improves MIoU by 1.29% and Mpa by 3.05%.
引用
收藏
页码:5774 / 5789
页数:16
相关论文
共 52 条
[1]  
AL-AMRI S S., 2010, International Journal of Computer Science Issues, (IIJCS), vol, V7, issue, P32
[2]  
Bruzzone L, 2014, REMOTE SENS DIGIT IM, V18, P127, DOI 10.1007/978-94-007-7969-3_9
[3]   Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation [J].
Chen, Liang-Chieh ;
Zhu, Yukun ;
Papandreou, George ;
Schroff, Florian ;
Adam, Hartwig .
COMPUTER VISION - ECCV 2018, PT VII, 2018, 11211 :833-851
[4]   Clustering based on eigenspace transformation - CBEST for efficient classification [J].
Chen, Yanlei ;
Gong, Peng .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2013, 83 :64-80
[5]   MFGAN: multi feature guided aggregation network for remote sensing image [J].
Chu, Shengguang ;
Li, Peng ;
Xia, Min .
NEURAL COMPUTING & APPLICATIONS, 2022, 34 (12) :10157-10173
[6]   Thermal infrared remote sensing of urban heat: Hotspots, vegetation, and an assessment of techniques for use in urban planning [J].
Coutts, Andrew M. ;
Harris, Richard J. ;
Thu Phan ;
Livesley, Stephen J. ;
Williams, Nicholas S. G. ;
Tapper, Nigel J. .
REMOTE SENSING OF ENVIRONMENT, 2016, 186 :637-651
[7]   Multiscale Location Attention Network for Building and Water Segmentation of Remote Sensing Image [J].
Dai, Xin ;
Xia, Min ;
Weng, Liguo ;
Hu, Kai ;
Lin, Haifeng ;
Qian, Ming .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2023, 61
[8]   Status of land cover classification accuracy assessment [J].
Foody, GM .
REMOTE SENSING OF ENVIRONMENT, 2002, 80 (01) :185-201
[9]   Application of Remote Sensing in Water Resource Management: The Case Study of Lake Trasimeno, Italy [J].
Giardino, Claudia ;
Bresciani, Mariano ;
Villa, Paolo ;
Martinelli, Angiolo .
WATER RESOURCES MANAGEMENT, 2010, 24 (14) :3885-3899
[10]   Optical remotely sensed time series data for land cover classification: A review [J].
Gomez, Cristina ;
White, Joanne C. ;
Wulder, Michael A. .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2016, 116 :55-72