收藏切换
A Method for Identifying Retrogressive Thaw Slumps Based on an Improved DeepLabV3 Model
收藏切换
Xueke MA1, 2, 3, Guanjun WEI1, 2, 3, *, Jie LI4, Zhigang DU5
Journal of Geodesy and Geodynamics | 2026, 46(9) : 1134 - 1146
Less
收藏切换
Journal of Geodesy and Geodynamics | 2026, 46(9): 1134-1146
A Method for Identifying Retrogressive Thaw Slumps Based on an Improved DeepLabV3 Model
Full
Xueke MA1, 2, 3, Guanjun WEI1, 2, 3, *, Jie LI4, Zhigang DU5
Affiliations
  • 1 Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou 730070, China
  • 2 National Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring, Lanzhou 730070, China
  • 3 Gansu Provincial Engineering Laboratory of Geographical Monitoring, Lanzhou 730070, China
  • 4 Employee Training Center of China Railway Lanzhou Bureau Group Co Ltd, Lanzhou 730050, China
  • 5 China Railway Tunnel Bureau Group Co Ltd, Guangzhou 511458, China
Published: 2026-09-15 doi: 10.14075/j.jgg.2025.12.440
Outline
收藏切换

This study proposes an improved DeepLabV3 model for the identification of retrogressive thaw slumps in frozen ground regions. The method integrates the Sobel operator with a boundary-enhanced dual attention (BEDA) module to accurately extract retrogressive thaw slump boundaries; a deformable context pyramid (DCP) module is employed to adapt to retrogressive thaw slumps of varying scales and morphological characteristics. Weighted residual connections are introduced to preserve the original semantic information and enhance detail representation. The dataset of retrogressive thaw slumps was constructed through visual interpretation using SBAS-InSAR surface deformation results, mountain shadow information, and Google Earth imagery. Experimental results show that the overall outputs of the model are closer to the ground truth labels, and the Recall and MIoU of the improved model are increased by 1.59% and 2.09%, respectively, compared with the baseline model.

retrogressive thaw slumps detection  /  deep learning  /  DeepLabV3  /  SBAS-InSAR  /  permafrost region
马雪珂, 魏冠军, 李婕, 杜志刚. 一种基于改进DeepLabV3模型的热融滑塌识别方法. 大地测量与地球动力学, 2026 , 46 (9) : 1134 -1146 . DOI: 10.14075/j.jgg.2025.12.440
Xueke MA, Guanjun WEI, Jie LI, Zhigang DU. A Method for Identifying Retrogressive Thaw Slumps Based on an Improved DeepLabV3 Model[J]. Journal of Geodesy and Geodynamics, 2026 , 46 (9) : 1134 -1146 . DOI: 10.14075/j.jgg.2025.12.440
Year 2026 volume 46 Issue 9
38
15
Cite this Article
BibTeX
Article Info
doi: 10.14075/j.jgg.2025.12.440
  • Receive Date:2025-12-15
  • Online Date:2026-10-30
  • Published:2026-09-15
Article Data
Affiliations
History
  • Received:2025-12-15
Funding
Affiliations
    1 Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou 730070, China
    2 National Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring, Lanzhou 730070, China
    3 Gansu Provincial Engineering Laboratory of Geographical Monitoring, Lanzhou 730070, China
    4 Employee Training Center of China Railway Lanzhou Bureau Group Co Ltd, Lanzhou 730050, China
    5 China Railway Tunnel Bureau Group Co Ltd, Guangzhou 511458, China
References
Share
https://castjournals.cast.org.cn/joweb/ddcl/EN/10.14075/j.jgg.2025.12.440
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT