收藏切换
Study on Precise Identification of Mine Surface Features Based on the Dual-Branch DB-TransUNet Model
收藏切换
Qi YAO1, 2, Shaodan LI1, 2, 3, Pengshuai XUE2, 3, Xiaolong ZHU1, 4, Yunxu YUAN1, 4, Yincui HU1, 2, 3
Mining Research and Development | 2026, 46(4) : 214 - 221
Less
收藏切换
Mining Research and Development | 2026, 46(4): 214-221
Study on Precise Identification of Mine Surface Features Based on the Dual-Branch DB-TransUNet Model
Full
Qi YAO1, 2, Shaodan LI1, 2, 3, Pengshuai XUE2, 3, Xiaolong ZHU1, 4, Yunxu YUAN1, 4, Yincui HU1, 2, 3
Affiliations
  • 1.Hebei Key Laboratory of Geological Resources, Environment Monitoring and Protection, Shijiazhuang, Hebei 050022, China
  • 2.School of Geographical Sciences, Hebei Normal University, Shijiazhuang, Hebei 050024, China
  • 3.Hebei Technological Innovation Center for Remote Sensing Identification of Environmental Changes, Shijiazhuang, Hebei 050024, China
  • 4.Hebei Institute of Geological Environment Monitoring, Shijiazhuang, Hebei 050022, China
Published: 2026-04-25
Outline
收藏切换

To address the issue of insufficient surface feature recognition accuracy in traditional segmentation models caused by complex vegetation cover and blurred surface feature boundaries in mining remote sensing images, a dual-branch DB-TransUNet model was proposed. This model innovatively constructed a dual-branch encoding architecture and integrated features from both branches using an adaptive weighted fusion (AWF) module. By combining this with a channel adaptive enhancement (CA) module to optimize feature weight allocation during the decoding process, it achieved focused detection of features in key regions and effective suppression of background noise. The experimental results show that incorporating the AWF and CA modules improved the mIoU of the model by 4.16% and 78.63%, respectively. The full DB-TransUNet model achieves excellent segmentation results of mIoU=0.903 7 and mDice=0.948 8 in complex remote sensing scenarios. Compared to mainstream network segmentation architectures, the DB-TransUNet model demonstrates superior performance across all indicators, achieving higher accuracy and stronger generalization capabilities in the recognition and detailed segmentation of complex surface features in mining environments. The findings validates the effectiveness of a strategy combining complementary multi-source features with weighted fusion in mitigating vegetation-induced interference in the identification of mine surface features, thereby providing a reliable technical solution for high-precision intelligent monitoring in open-pit mining areas.

Mining remote sensing image  /  Surface feature recognition  /  Feature fusion  /  Dual-branch network  /  Vegetation interference
姚奇, 李少丹, 薛彭帅, 朱小龙, 袁运许, 胡引翠. 基于双分支DB-TransUNet模型的矿山地物精准识别研究. 矿业研究与开发, 2026 , 46 (4) : 214 -221 .
Qi YAO, Shaodan LI, Pengshuai XUE, Xiaolong ZHU, Yunxu YUAN, Yincui HU. Study on Precise Identification of Mine Surface Features Based on the Dual-Branch DB-TransUNet Model[J]. Mining Research and Development, 2026 , 46 (4) : 214 -221 .
Year 2026 volume 46 Issue 4
30
13
Cite this Article
BibTeX
Article Info
  • Receive Date:2025-05-12
  • Online Date:2026-10-29
  • Published:2026-04-25
Article Data
Affiliations
History
  • Received:2025-05-12
Funding
Affiliations
    1.Hebei Key Laboratory of Geological Resources, Environment Monitoring and Protection, Shijiazhuang, Hebei 050022, China
    2.School of Geographical Sciences, Hebei Normal University, Shijiazhuang, Hebei 050024, China
    3.Hebei Technological Innovation Center for Remote Sensing Identification of Environmental Changes, Shijiazhuang, Hebei 050024, China
    4.Hebei Institute of Geological Environment Monitoring, Shijiazhuang, Hebei 050022, China
References
Share
https://castjournals.cast.org.cn/joweb/kyyjykf/EN/
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