Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet

Fei Sun; Run Wang; Xueming Li; Yurong Guan

doi:10.47297/taposatWSP2633-456908.20250610

PDF
导出
分享
收藏
专辑

Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet

6卷第10期

浏览量: 0 Downloads: 0

Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet
Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet
1.Huanggang Normal University, School of Computer Science and Artificial Intelligence, Huanggang, Hubei 438000, China
2.Hubei Key Laboratory of Resources and Eco-environmental Geology, Wuhan, Hubei, 430034, China
3.Hubei Geological Environment Station, Wuhan, Hubei, 430034, China
4.Huanggang Normal University, School of Foreign Studies, Huanggang, Hubei, 438000, China
2025年6卷第10期页码：37-42
纸质出版：2025-10-30
DOI：10.47297/taposatWSP2633-456908.20250610
Accepted：

Scan QR Code

Fei Sun, Run Wang, Xueming Li & Yurong Guan. (2025). Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet. 科技理论与实践（英文版）, 6(10), 37-42.

Sun Fei,Wang Run,Li Xueming,et al.Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet[J].Theory and Practice of Science and Technology,2025,06(10):37-42.
Fei Sun, Run Wang, Xueming Li & Yurong Guan. (2025). Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet. 科技理论与实践（英文版）, 6(10), 37-42. DOI： 10.47297/taposatWSP2633-456908.20250610.

Sun Fei,Wang Run,Li Xueming,et al.Research on Abstract Semantic Segmentation of Remote Sensing Data Based on UPerNet[J].Theory and Practice of Science and Technology,2025,06(10):37-42. DOI： 10.47297/taposatWSP2633-456908.20250610.

摘要

Abstract

In practical production and life scenarios

higher-level abstract semantics are typically of greater concern. This paper

based on deep learning methods

uses the abstract semantics of mining land degradation as an example and investigates the feasibility of directly segmenting high-level abstract semantics from GF-2 remote sensing data using the UPerNet model. Experimental studies on mining area data from three typical landform features in Hubei Province demonstrate that: 1) For abstract semantic classification of mining damage types

the UPerNet model achieves efficient and accurate segmentation

outperforming the U-Net model under identical experimental conditions; 2) In cases with limited and imbalanced datasets

UPerNet exhibits superior generalization ability when dealing with less distinguishable abstract semantics. These results confirm the feasibility of using the UPerNet model to extract higher-order abstract semantics from remote sensing appearance features

providing technical support for investigating mining land degradation and offering a reference for directly extracting high-level abstract semantics from remote sensing data in complex application scenarios.

关键词

Keywords

references

General Office of the Ministry of Natural Resources. Notice of the General Office of the Ministry of Natural Resources on carrying out the pilot work for the survey of land damaged by mining across the country[Z]. Beijing, 2024-07-18.

Xiao S. Identification and Location Method of Sedimentation Tank in High Resolution Remote Sensing Image Based on the Improved Yolov3 Model [D]. Jiangxi University of Science and Technology, 2021.

Li H, Xiao S, Wang X, et al . High-resolution remote sensing image rare earth mining identification method based on Mask R-CNN [J ] . Journal of China University of Mining & Technology, 2020, 49 (06): 1215-1222.

Zhang C, Li F, Li J, et al . Recognition of land use on open-pit coal mining area based on DeepLabv3+ and GF-2 high-resolution images [J ] . Coal Geology & Exploration,2022, 50 (06): 94-103.

Song R, Zheng H, Wang D, et al . Classification of features in open pit mining areas based on deep learning and high resolution remote sensing images [J ] . China Mining Magazine,2022, 31 (07): 102-111.

Zhu C, Chen Y, Lin J, et al . Empirical Study on Land Use Transformation of Rare Earth Mining Area in Hilly and Mountainous Areas of South China [J ] . Journal of Gannan Normal University, 2022, 43 (06): 108-115.

Jiao X. Research on Unstructured Road Recognition and Lane Departure Detection in Mining Area based on Machine Vision [D]. Xi'an University of Architecture and Technology, 2022.

Zhao Y, Wei H, Dong J, et al . Opencast coal mine scene recognition based on sub-region multi-label learning [J ] . National Remote Sensing Bulletin, 2022, 26 (09): 1849-1858.

Ruan S, Jing W, Ji ng Y, et al . Detection on Vegetation Coverage Changes in Mining Areas Based on Improved DeepLabV3+[J ] . Mining R & D,2023, 43 (07): 143-151.

Fan S, Gong M, Liu J, et al . Identification of Surface Hazards in Mining Areas Based on DeeplabV3+ Network and Attention Mechanism: Taking 30507 Working Face of the Tashan Coal Mine as an Example [J ] . Science Technology and Industry, 2024, 24 (17): 278-286.

Yuan X, Shi J, Gu L. A review of deep learning methods for semantic segmentation of remote sensing imagery[J]. Expert Systems with Applications, 2021, 169: 114417.

Cheng J, Deng C, Su Y, et al . Methods and datasets on semantic segmentation for Unmanned Aerial Vehicle remote sensing images: A review[J ] . ISPRS Journal of Photogrammetry and Remote Sensing,2024, 211: 1-34.

Sun F, Fang F, Wang R, et al . An Impartial Semi-Supervised Learning Strategy for Imbalanced Classification on VHR Images[J ] .Sensors, 2020, 20 (22).

Xiao T, Liu Y, Zhou B, et al . Unified Perceptual Parsing for Scene Understanding[C ] . Computer Vision(ECCV 2018). Cham: Springer, 2018: 432-448.

Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation[C]. Medical Image Computing and Computer-Assisted Intervention (MICCAI 2015). Cham: Springer, 2015: 234-241.

Russakovsky O, Deng J, Su H, et al . ImageNet Large Scale Visual Recognition Challenge [J ] . International Journal of Computer Vision, 2015, 115 (3): 211-252.

Lin T Y, Goyal P, Girshick R, et al. Focal Loss for Dense Object Detection[C].2017 IEEE International Conference on Computer Vision (ICCV). Venice: IEEE, 2017: 2999-3007.

Dong J. Focal Loss Improves the Model Performance on Multi-Label Image Classifications with Imbalanced Data[C]. Proceedings of the 2nd International Conference on Industrial Control Network and System Engineering Research. Kuala Lumpur, Malaysia: Association for Computing Machinery, 2020: 18–21.

文章被引用时，请邮件提醒。

Submit

Energy Systems Research Based on Algorithm Optimization and Intelligent Management

Research on Game Anti-Cheating Methods Based on Behavioral Analysis and Artificial Intelligence

Advancing License Plate Recognition in Complex Environments: A Hybrid Approach Integrating Deep Learning and Image Pre-Processing

A Lightweight Real-time Detection System for Power Line Defects Based on DartNet-YOLO

Add Sharpening to the Super-Resolution Image Processing

Related Author

Zihang Hong School of Electrical Engineering, Shangluo University

Borui Li Kelowna Christian School, Kelowna V1W3G7

Siyuan Zhang School of Engineering, Royal Melbourne Institute of Technology;College of Civil Aviation NUAA, Nanjing University of Aeronautics and Astronautics

Zitong Xu Jinan Xinhang Experimental Foreign Language School

Zhixuan Wang School of Automation, Wuhan University of Technology

Kazem Chamran City University Malaysia, SintokKedah Darul Aman

Yuefu Qiu City University Malaysia, SintokKedah Darul Aman

Kehan Li School of Information Engineering, Sichuan Agricultural University

Related Institution

School of Electrical Engineering, Shangluo University

Kelowna Christian School, Kelowna V1W3G7

College of Civil Aviation NUAA, Nanjing University of Aeronautics and Astronautics

School of Engineering, Royal Melbourne Institute of Technology

Jinan Xinhang Experimental Foreign Language School

⁰