面向稀土矿区高光谱精细分类的多层注意力卷积神经网络模型

doi:10.3964/j.issn.1000-0593(2025)09-2666-10

Abstract
References
Related Citation (15)

Download: PDF (23066 KB)
Export: BibTeX | EndNote (RIS)

Abstract Ion-adsorption-type rare earth minerals are important strategic resources. Long-term extensive mining has led to severe surface damage in mining areas, posing significant challenges to the ecological environment. Accurate and detailed land use information is a critical foundation for ecological restoration and process monitoring in mining areas. Hyperspectral imagery is considered an effective means for large-scale monitoring of mining areas to obtain land use information. However, the complexity of the land cover and the information redundancy in hyperspectral images pose challenges for fine classification. This study proposes a fine classification method for rare earth mining areas based on object-oriented thinking and a multi-layer attention convolutional neural network (OCTC). First, a scale parameter estimation model was used to quantitatively analyze the optimal segmentation scale at multiple levels of the rare earth mining area images. Four types of image features—spectral, index, texture, and geometric—were extracted from the segmented images. Then, an optimal feature combination was obtained through distance separability analysis. Based on this, a multi-layer attention convolutional neural network model (OCTC) was used for classification. This model is an improved version of the 1D-CNN, integrating the Transformer and CBAM to enhance the model's feature extraction capabilities and overall classification accuracy. To verify the method's effectiveness, Zhuhai-1 hyperspectral remote sensing imagery was used as the data source, and the Jiangxi Gan'nan Lingbei rare earth mining area served as the study region. The proposed method was compared with KNN, RF, and 1D-CNN classification methods for accuracy analysis. The results demonstrate that the proposed method effectively mitigates salt-and-pepper noise, maintains good overall classification integrity, and achieves the highest classification accuracy. The overall accuracy reached 88.11%, representing an improvement of 1.22% to 8.84% compared to other classification methods, with the Kappa coefficient increasing by 0.015 9 to 0.109 0. This method can provide valuable reference and scientific insights for the fine classification of land use and production monitoring, as well as environmental protection management in rare earth mining areas.

Key words：Object-oriented convolutional neural network; Zhuhai-1; Hyperspectral remote sensing; Ion-adsorption rare earth; Land use

Received: 2024-12-18 Accepted: 2025-04-12

ZTFLH:

TP751

Corresponding Authors: LI Heng-kai E-mail: giskai@jxust.edu.cn

Cite this article:

FAN Xiao-yong,LI Heng-kai,LIU Kun-ming, et al. A Multi-Layer Attention Convolutional Neural Network Model for Fine Classification of Hyperspectral Images in Rare Earth Mining Areas[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(09): 2666-2675.

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2025)09-2666-10 OR https://www.gpxygpfx.com/EN/Y2025/V45/I09/2666

[1] Wang J L，Guo M Y，Liu M M，et al. Resources Policy，2020，65：101569.
[2] LIU Yu-jiu（刘余九）. Journal of the Chinese Society of Rare Earths（中国稀土学报），2007，25(3)：257.
[3] Zhang J Y，Li H K，Long B P，et al. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2024，17：10726.
[4] Li Y S，Li H K，Pan J. Restoration Ecology，2023，32(2)：e13994.
[5] SU Han, CHEN Na, PENG Jiang-tao，et al（苏涵，陈娜，彭江涛，等）. National Remote Sensing Bulletin（遥感学报），2024，28(1)：247.
[6] Chen W，Li X，He H，et al. Remote Sensing，2018，10(1)：15.
[7] Mao Z J，Yu H Y，Liang W，et al. Ecological Engineering，2024，207: 107304.
[8] FAN Pei-wei，HAO Meng-meng，JIANG Dong，et al（范沛薇，郝蒙蒙，江东，等）. Remote Sensing Technology and Application（遥感技术与应用），2023，38(2)：274.
[9] Wu Z B，Li H K，Wang Y Q，et al. Expert Systems with Applications，2024，247：123233.
[10] LIU Yan-jun，LIU Kai，CAO Jing-jing（刘燕君，刘凯，曹晶晶）. Bulletin of Surveying and Mapping（测绘通报），2023，12：136.
[11] ZHANG Bing（张兵）. National Remote Sensing Bulletin（遥感学报），2016，20(5)：1062.
[12] Li C G，Li H K，Zhou Y B，et al. Sustainability，2024，16(9)：3582.
[13] HUO Guang-jie，HU Nai-xun，CHEN Tao，et al（霍光杰，胡乃勋，陈涛，等）. Journal of Henan Polytechnic University (Natural Science)[河南理工大学学报（自然科学版）]，2021，40(2)：70.
[14] Zhou G D，Xu J H，Chen W T，et al. IEEE Transactions on Geoscience and Remote Sensing，2023，61：4401220.
[15] ZHANG Cheng-ye，LI Jun，LEI Shao-gang，et al（张成业，李军，雷少刚，等）. Metal Mine（金属矿山），2022，3：1.
[16] LU Chun-yan，LEI Yi-fan，SU Ying，et al（路春燕，雷依凡，苏颖，等）. Remote Sensing Technology and Application（遥感技术与应用），2021，36(4)：713.
[17] Hao S，Cui Y，Wang J. Sensors，2021，21(23)：7935.
[18] Guo Q，Zhang J，Guo S J，et al. Remote Sensing，2022，14：3885.
[19] HUANG Yu-fei，LU Chun-yan，JIA Ming-ming，et al（黄雨菲，路春燕，贾明明，等）. Biodiversity Science（生物多样性），2023，31(3)：22411.
[20] Drägut L，Tiede D，Levick S R. International Journal of Geographical Information Science，2010，24(6)：859.
[21] Chao X W，Li Y. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2022，15：8798.
[22] Pal M. International Journal of Remote Sensing，2005，26(1)：217.
[23] Ye F，Zhou B. Sensors，2024，24: 4108.
[24] Balha A，Mallick J，Pandey S，et al. Earth Science Informatics，2021，14(4)：2231.
[25] Zoleikani R，Valadan Zoej M J，Mokhtarzadeh M. Journal of the Indian Society of Remote Sensing，2017，45(1)：25.
[26] Zhao J L，Wu K K，Zhang L，et al. Signal Processing，2025，230：109850.
[27] Paoletti M E，Haut J M，Plaza J，et al. ISPRS Journal of Photogrammetry and Remote Sensing，2019，158：279.