Build and application of automatic classification model of land damage in the middle section of Tianshan Mountains
-
摘要:
新疆天山中段矿产资源丰富，但高强度采矿活动导致土地损毁问题日益加剧。针对矿山土地损毁监测效率低、传统遥感解译依赖人工经验等问题，文章提出了基于神经网络的遥感影像土地损毁自动分类模型——SENetV2-COT-DeepLabV3+，该模型是在DeepLabV3+模型基础上，融合了上下文转换器模块与SENetV2模块，从而增强了上下文特征提取和通道注意力机制能力，优化了模型对复杂矿山地物的分割能力。根据高分系列遥感影像，构建了包含
59198 个样本的天山中段矿山样本集，通过数据增强扩展至177594 个样本；使用该数据训练SENetV2-COT-DeepLabV3+模型，提高其泛化能力与识别精度，精准掌握矿产资源开发造成的土地损毁分布和程度；通过与FCN、UNeT、PSPNeT等模型进行对比试验得出该改进模型在平均交并比、平均召回率、平均精确率和平均系数等4项指标上均优于FCN、PSPNet等主流模型，分割精度较DeepLabV3+提升了1.63%～2.34%。基于该模型在pycharm平台搭建了矿区土地损毁类型深度学习遥感解译系统，目前该系统已部署至当地矿山管理部门，识别准确率达85%以上，实现了高精度、高效率的土地损毁识别，为矿区土地损毁动态监测与生态修复管理提供了智能化解决方案，推动矿山开发与环境保护的协调发展。Abstract:The middle section of the Tianshan Mountains in Xinjiang is rich in mineral resources, but the problem of land damage has been exacerbated by high-intensity mining activities. In order to solve the problems of low efficiency of mine land damage monitoring and traditional remote sensing interpretation relying on manual experience, this paper proposes an automatic classification model of remote sensing image land damage based on neural network-SENetV2-COT-DeepLabV3, which is based on the DeepLabV3 model and integrates the Contextual Transformer (COT) module and the SENetV2 module, so as to enhance the ability of context feature extraction and channel attention mechanism. Optimize the model's segmentation ability for complex mine features. Firstly, according to the high-resolution series of remote sensing images, a sample set of
59198 samples was constructed for the middle section of the Tianshan Mountains, which was extended to177594 samples through data augmentation. Then, the SENetV2-COT-DeepLabV3 model was trained to improve its generalization ability and recognition accuracy, and accurately grasp the distribution and extent of land damage caused by mineral resource development. Finally, through comparative experiments with FCN, UNeT, PSPNeT and other models, it is concluded that the improved model is better than the mainstream models such as FCN and PSPNet in four indicators: MIoU, mRecall, mPrecision, and mDice, and the segmentation accuracy is 1.63%−2.34% higher than that of DeepLabV3. Based on the model, a deep learning remote sensing interpretation system for land damage types in mining areas was built on the Pycharm platform, which has been deployed to the local mine management department, with a recognition accuracy of more than 85%, realizing high-precision and high-efficiency land damage identification, providing an intelligent solution for dynamic monitoring and ecological restoration management of land damage in mining areas, and promoting the coordinated development of mine development and environmental protection. -
-
表 1 主要卫星数据参数
Table 1. Main satellite data parameters
影像名称 发射时间 影像分辨率 高分1号影像 2013-04-26 全色2 m，多光谱8 m 高分2号影像 2014-08-09 全色0.8 m，多光谱3.2 m 高分7号影像 2018-06-02 全色2 m，多光谱8 m 资源3号影像 2019-11-03 全色0.8 m，多光谱2.6 m 注：遥感影像数据来自新疆维吾尔自治区测绘成果中心（国家卫星测绘应用中心新疆分中心）。 
下载: 导出CSV
表 2 天山中段矿山地物影像特征
Table 2. Image features of mine objects in the middle of Tianshan Mountains
地物
类型解译标志 影像特征 植被 
矿区周围植被通常集中分布，在影像上呈现出浅绿色、深绿色丝绒状纹理 水域 
水域在影像上呈现蓝色、深绿色，矿区内水域形状差异较大 生活区 
生活区建筑较为集中，形状规整通常表现为矩形，生活区内具有道路、绿化等，屋顶颜色上以灰白色、蓝色
为主工业
广场
工业广场有明显的结构布局形状规整多为矩形或方形，具有水泥路面，在影像上呈现蓝色和高亮的灰
白色堆放场 
堆放场在影像上有着明显凸起，形状多为圆形、椭圆形、圆扇形，颜色为灰色 尾矿库 
尾矿库形状较为规则，地界明显，内存在积水，水体随着尾砂浓度由浅色逐渐接近自然水体颜色，尾矿库表面覆盖薄膜在影像中呈现高亮色调 露天
采场
与周围地貌相比有着明显破坏痕迹，呈负地形，采场周围伴随着阶梯状纹理，边界清晰，采场形状多为不规则形体并在颜色上呈现高亮色调，周边无植被发育
下载: 导出CSV
表 3 天山中段矿山数据集
Table 3. Data sets of mines in the middle of Tianshan Mountains
地物类别 样本数量 数据增强 训练集（60%） 验证集（20%） 测试集（20%） 植被 5556 16668 10000 3334 3334 水域 4298 12894 7736 2579 2579 生活区 8196 24588 14752 4918 4918 工业广场 21862 65586 39352 13117 13117 堆放场 10524 31572 18944 6314 6314 露天采场 8628 25884 15530 5177 5177 尾矿库 134 402 242 80 80 总计 59198 177594 106556 35519 35519
下载: 导出CSV
表 4 消融试验结果
Table 4. Ablation experimental results
分割模型 MIoU/% mRecall/% mPrecision/% mDice/% DL 86.04 84.42 95.73 88.97 COT-DL 87.43 85.29 95.94 91.12 SE-DL 86.92 84.51 95.85 89.78 SE-COT-DL 87.67 85.95 96.44 91.31
下载: 导出CSV
表 5 模型算法试验数据结果
Table 5. Comparison results of experimental data of mine algorithm in the middle section of the Tianshan Mountains
分割模型 MIoU/% mRecall/% mPrecision/% mDice/% FCN 63.02 51.02 66.92 58.10 UNeT 65.23 53.84 67.4 62.49 PSPNeT 85.16 81.98 93.02 88.94 DeepLabV3+ 86.04 83.42 95.73 88.97 SE-DL 86.92 83.51 95.85 89.78 SE-COT-DL 87.67 85.95 96.44 91.31
下载: 导出CSV
-
[1] 胡振琪. 我国土地复垦与生态修复30年:回顾、反思与展望[J]. 煤炭科学技术,2019,47(1):25 − 35. [HU Zhenqi. The 30 years’ land reclamation and ecological restoration in China:Review,rethinking and prospect[J]. Coal Science and Technology,2019,47(1):25 − 35. (in Chinese with English abstract)]
HU Zhenqi. The 30 years’ land reclamation and ecological restoration in China: Review, rethinking and prospect[J]. Coal Science and Technology, 2019, 47（1）: 25 − 35. （in Chinese with English abstract）
[2] 童心仪. 面向广域土地覆盖的高分辨率遥感影像分类方法研究[D]. 武汉:武汉大学,2020. [TONG Xinyi. Large-scale land-cover classification with high-resolution remote sensing images[D]. Wuhan:Wuhan University,2020. (in Chinese with English abstract)]
TONG Xinyi. Large-scale land-cover classification with high-resolution remote sensing images[D]. Wuhan: Wuhan University, 2020. （in Chinese with English abstract）
[3] 杨辉. 基于视觉注意力机制深度学习模型的高分辨率遥感影像信息提取研究[D]. 武汉:武汉大学,2019. [YANG Hui. The research of information extraction from VHR image based on deep learning model with attention mechanism[D]. Wuhan:Wuhan University,2019. (in Chinese with English abstract)]
YANG Hui. The research of information extraction from VHR image based on deep learning model with attention mechanism[D]. Wuhan: Wuhan University, 2019. （in Chinese with English abstract）
[4] 李丽,杨金中,陈栋,等. 长江经济带江苏段废弃露天矿山分布与生态修复遥感调查研究[J]. 水文地质工程地质,2022,49(1):183 − 190. [LI Li,YANG Jinzhong,CHEN Dong,et al. Remote sensing investigation and study on the distribution and ecological restoration of abandoned open-pit mines in Jiangsu section of the Yangtze River Economic Belt[J]. Hydrogeology & Engineering Geology,2022,49(1):183 − 190. (in Chinese with English abstract)]
LI Li, YANG Jinzhong, CHEN Dong, et al. Remote sensing investigation and study on the distribution and ecological restoration of abandoned open-pit mines in Jiangsu section of the Yangtze River Economic Belt[J]. Hydrogeology & Engineering Geology, 2022, 49（1）: 183 − 190. （in Chinese with English abstract）
[5] 刘仁钊,廖文峰. 遥感图像分类应用研究综述[J]. 地理空间信息,2005,3(5):11 − 13. [LIU Renzhao,LIAO Wenfeng. A review of research on the application of remote sensing image classification[J]. Geospatial Information,2005,3(5):11 − 13. (in Chinese with English abstract)]
LIU Renzhao, LIAO Wenfeng. A review of research on the application of remote sensing image classification[J]. Geospatial Information, 2005, 3（5）: 11 − 13. （in Chinese with English abstract）
[6] 杨显华,黄洁,田立,等. 四川省矿山遥感监测主要成果与进展[J]. 中国地质调查,2016,3(5):41 − 47. [YANG Xianhua,HUANG Jie,TIAN Li,et al. Major achievements and progress of remote sensing monitoring mines in Sichuan Province[J]. Geological Survey of China,2016,3(5):41 − 47. (in Chinese with English abstract)]
YANG Xianhua, HUANG Jie, TIAN Li, et al. Major achievements and progress of remote sensing monitoring mines in Sichuan Province[J]. Geological Survey of China, 2016, 3（5）: 41 − 47. （in Chinese with English abstract）
[7] 彭瑛,张志,彭丹,等. 鄂西聚磷区矿产开发多目标遥感监测目视解译技术[J]. 工矿自动化,2009,35(4):6 − 9. [PENG Ying,ZHANG Zhi,PENG Dan,et al. Visual interpretation technology of multi-target remote monitoring for mineral development in gather phosphorus area of western of Hubei Province[J]. Industry and Mine Automation,2009,35(4):6 − 9. (in Chinese with English abstract)]
PENG Ying, ZHANG Zhi, PENG Dan, et al. Visual interpretation technology of multi-target remote monitoring for mineral development in gather phosphorus area of western of Hubei Province[J]. Industry and Mine Automation, 2009, 35（4）: 6 − 9. （in Chinese with English abstract）
[8] 包玉斌,姚建华,黄涛,等. 银川市国土空间土地利用变化多源遥感人工目视解译分析[J]. 安徽农业科学,2021,49(16):221 − 229. [BAO Yubin,YAO Jianhua,HUANG Tao,et al. Analysis of land use change of national land space in Yinchuan based on manual visual interpretation of multi-source remote sensing images[J]. Journal of Anhui Agricultural Sciences,2021,49(16):221 − 229. (in Chinese with English abstract)]
BAO Yubin, YAO Jianhua, HUANG Tao, et al. Analysis of land use change of national land space in Yinchuan based on manual visual interpretation of multi-source remote sensing images[J]. Journal of Anhui Agricultural Sciences, 2021, 49（16）: 221 − 229. （in Chinese with English abstract）
[9] 陈树彪. 遥感影像判读在森林资源二类调查中的应用与研究——以西林吉林业局判读调查工作为例[J]. 内蒙古林业调查设计,2016,39(2):16 − 17. [CHEN Shubiao. The application and research of remote sensing image interpretation in the second-class investigation of forest resources:Taking the interpretation and investigation work of Xilinji Forestry Bureau as an example[J]. Inner Mongolia Forestry Investigation and Design,2016,39(2):16 − 17. (in Chinese with English abstract)]
CHEN Shubiao. The application and research of remote sensing image interpretation in the second-class investigation of forest resources: Taking the interpretation and investigation work of Xilinji Forestry Bureau as an example[J]. Inner Mongolia Forestry Investigation and Design, 2016, 39（2）: 16 − 17. （in Chinese with English abstract）
[10] 于晨. 典型矿山环境遥感地物智能识别方法的研究[D]. 北京:中国地质大学(北京),2020. [YU Chen. Research on intelligent recognition method of remote sensing features in typical mine environment[D]. Beijing:China University of Geosciences(Beijing),2020. (in Chinese with English abstract)]
YU Chen. Research on intelligent recognition method of remote sensing features in typical mine environment[D]. Beijing: China University of Geosciences（Beijing）, 2020. （in Chinese with English abstract）
[11] 黄登冕. 基于深度学习的典型矿山地物智能识别方法研究——以祁连山为例[D]. 兰州:西北师范大学,2022. [HUANG Dengmian. Research on intelligent recognition method of typical mine features based on deep learning:A case study of the Oilian Mountains[D]. Lanzhou:Northwest Normal University,2022. (in Chinese with English abstract)]
HUANG Dengmian. Research on intelligent recognition method of typical mine features based on deep learning: A case study of the Oilian Mountains[D]. Lanzhou: Northwest Normal University, 2022. （in Chinese with English abstract）
[12] 杜岩,霍磊晨,张洪达,等. 岩块体崩塌灾害遥感监测预警试验研究[J]. 中国矿业大学学报,2022,51(6):1201 − 1208. [DU Yan,HUO Leichen,ZHANG Hongda,et al. Experimental research on remote monitoring and early warning of rock collapse[J]. Journal of China University of Mining & Technology,2022,51(6):1201 − 1208. (in Chinese with English abstract)]
DU Yan, HUO Leichen, ZHANG Hongda, et al. Experimental research on remote monitoring and early warning of rock collapse[J]. Journal of China University of Mining & Technology, 2022, 51（6）: 1201 − 1208. （in Chinese with English abstract）
[13] 何金宝,孔繁鹏,裴艳东,等. 基于递归特征消除与贝叶斯优化的盐碱地信息遥感提取模型研究[J/OL]. 水文地质工程地质,2025. (2025-04-30)[2025-05-05]. [HE Jinbao,KONG Fanpeng,PEI Yandong,et al. Research on remote sensing extraction model of saline-alkali land information based on recursive feature cancellation and Bayesian optimization[J/OL]. Hydrogeology & Engineering Geology,2025. (2025-04-30)[2025-05-05]. http://kns.cnki.net/kcms/detail/11.2202.P.20250429.1819.002.html. (in Chinese with English abstract)]
HE Jinbao, KONG Fanpeng, PEI Yandong, et al. Research on remote sensing extraction model of saline-alkali land information based on recursive feature cancellation and Bayesian optimization[J/OL]. Hydrogeology & Engineering Geology, 2025. （2025-04-30）[2025-05-05]. http://kns.cnki.net/kcms/detail/11.2202.P.20250429.1819.002.html. （in Chinese with English abstract）
[14] 吴虹,杨永德,王松庆. Quickbird-2&SPOT-1矿山生态环境遥感调查试验研究[J]. 国土资源遥感,2004,16(4):46 − 49. [WU Hong,YANG Yongde,WANG Songqing. The tentative application of quickbird-2 and spot-1 satellite remote sensing to the ecological environment investigation of mines[J]. Remote Sensing for Land & Resources,2004,16(4):46 − 49. (in Chinese with English abstract)]
WU Hong, YANG Yongde, WANG Songqing. The tentative application of quickbird-2 and spot-1 satellite remote sensing to the ecological environment investigation of mines[J]. Remote Sensing for Land & Resources, 2004, 16（4）: 46 − 49. （in Chinese with English abstract）
[15] 漆小英,杨武年,邵怀勇,等. 矿山扩展遥感信息自动提取方法研究——以攀枝花钒钛磁铁矿为例[J]. 测绘科学,2008,33(3):76 − 78. [QI Xiaoying,YANG Wunian,SHAO Huaiyong,et al. An automatic extraction method of mining area expanding remote sensing image:A study on Panzhihua vanadium titanium magnetite ore[J]. Science of Surveying and Mapping,2008,33(3):76 − 78. (in Chinese with English abstract)] doi: 10.3771/j.issn.1009-2307.2008.03.026
QI Xiaoying, YANG Wunian, SHAO Huaiyong, et al. An automatic extraction method of mining area expanding remote sensing image: A study on Panzhihua vanadium titanium magnetite ore[J]. Science of Surveying and Mapping, 2008, 33（3）: 76 − 78. （in Chinese with English abstract） doi: 10.3771/j.issn.1009-2307.2008.03.026
[16] 何亮柱,洪金益,张建国,等. 基于IKONOS数据的矿山开发占地信息提取[J]. 国土资源遥感,2013,25(1):150 − 154. [HE Liangzhu,HONG Jinyi,ZHANG Jianguo,et al. Research on extraction of land information used by mining development based on IKONOS remote sensing data[J]. Remote Sensing for Land & Resources,2013,25(1):150 − 154. (in Chinese with English abstract)]
HE Liangzhu, HONG Jinyi, ZHANG Jianguo, et al. Research on extraction of land information used by mining development based on IKONOS remote sensing data[J]. Remote Sensing for Land & Resources, 2013, 25（1）: 150 − 154. （in Chinese with English abstract）
[17] 朱青,林建平,国佳欣,等. 基于影像特征CART决策树的稀土矿区信息提取与动态监测[J]. 金属矿山,2019(5):161 − 169. [ZHU Qing,LIN Jianping,GUO Jiaxin,et al. Information extraction and dynamic monitoring of rare earth mining area based on image feature CART decision tree[J]. Metal Mine,2019(5):161 − 169. (in Chinese with English abstract)]
ZHU Qing, LIN Jianping, GUO Jiaxin, et al. Information extraction and dynamic monitoring of rare earth mining area based on image feature CART decision tree[J]. Metal Mine, 2019（5）: 161 − 169. （in Chinese with English abstract）
[18] JIANG Jiahuan,FU Xiongjun,QIN Rui,et al. High-speed lightweight ship detection algorithm based on YOLO-v4 for three-channels RGB SAR image[J]. Remote Sensing,1909,13(10):1909.
[19] 王民水,孔祥明,陈学业,等. 基于随机补片和DeepLabV3+的建筑物遥感图像变化检测[J]. 吉林大学学报(地球科学版),2021,51(6):1932 − 1938. [WANG Minshui,KONG Xiangming,CHEN Xueye,et al. Remote sensing image change detection based on random patches and DeepLabV3+ network[J]. Journal of Jilin University(Earth Science Edition),2021,51(6):1932 − 1938. (in Chinese with English abstract)]
WANG Minshui, KONG Xiangming, CHEN Xueye, et al. Remote sensing image change detection based on random patches and DeepLabV3+ network[J]. Journal of Jilin University（Earth Science Edition）, 2021, 51（6）: 1932 − 1938. （in Chinese with English abstract）
[20] 葛振强. 基于DeepLabV3+模型的工地裸露土地图像语义分割研究与实现[D]. 太原:太原师范学院,2024. [GE Zhenqiang. Research and implementation of construction site exposed soil image semantic segmentation based on the DeepLabV3+ Model[D]. Taiyuan:Taiyuan Normal University,2024. (in Chinese with English abstract)]
GE Zhenqiang. Research and implementation of construction site exposed soil image semantic segmentation based on the DeepLabV3+ Model[D]. Taiyuan: Taiyuan Normal University, 2024. （in Chinese with English abstract）
[21] 马小淞,闫庆武,移明昊,等. 天山中段城市化水平与生态环境质量时空耦合研究[J]. 生态与农村环境学报,2025,41(2):172 − 182. [MA Xiaosong,YAN Qingwu,YI Minghao,et al. Research on the spatio-temporal coupling of urbanization and ecological environment quality in the middle section of the Tianshan Mountains[J]. Journal of Ecology and Rural Environment,2025,41(2):172 − 182. (in Chinese with English abstract)]
MA Xiaosong, YAN Qingwu, YI Minghao, et al. Research on the spatio-temporal coupling of urbanization and ecological environment quality in the middle section of the Tianshan Mountains[J]. Journal of Ecology and Rural Environment, 2025, 41（2）: 172 − 182. （in Chinese with English abstract）
[22] LI M S. Ecological restoration of mineland with particular reference to the metalliferous mine wasteland in China:A review of research and practice[J]. Science of the Total Environment,2006,357(1/2/3):38 − 53.
[23] 吕杰. 尾矿库遥感监测技术研究与应用——以河北省尾矿库为例[D]. 北京:中国地质大学(北京),2014. [LV Jie. Remote sensing research and application of tailings pond:A case study of tailings pond in Hebei Province[D]. Beijing:China University of Geosciences (Beijing),2014. (in Chinese with English abstract)]
LV Jie. Remote sensing research and application of tailings pond: A case study of tailings pond in Hebei Province[D]. Beijing: China University of Geosciences （Beijing）, 2014. （in Chinese with English abstract）
[24] 贾倩,刘彬彬,於方,等. 我国尾矿库突发环境事件统计分析与对策建议[J]. 安全与环境工程,2015,22(2):92 − 96. [JIA Qian,LIU Binbin,YU Fang,et al. Statistics analysis and management suggestion for the emergent environmental accident of tailing pond in China[J]. Safety and Environmental Engineering,2015,22(2):92 − 96. (in Chinese with English abstract)]
JIA Qian, LIU Binbin, YU Fang, et al. Statistics analysis and management suggestion for the emergent environmental accident of tailing pond in China[J]. Safety and Environmental Engineering, 2015, 22（2）: 92 − 96. （in Chinese with English abstract）
[25] 董兆祥,王颜祺,张明义. 矿区环境地质问题的预测与防治[M]. 北京:地质出版社,1997. [DONG Zhaoxiang,WANG Yanqi,ZHANG Mingyi. Prediction and prevention of environmental geological problems in mining area[M]. Beijing:Geological Publishing House,1997. (in Chinese)]
DONG Zhaoxiang, WANG Yanqi, ZHANG Mingyi. Prediction and prevention of environmental geological problems in mining area[M]. Beijing: Geological Publishing House, 1997. （in Chinese）
[26] 胡振琪. 矿山土地复垦与生态修复领域“十四五”高质量发展的若干思考[J]. 智能矿山,2021,2(1):29 − 32. [HU Zhenqi. Some thoughts on the high-quality development of the 14th five-year plan in the field of mine land reclamation and ecological restoration[J]. Journal of Intelligent Mine,2021,2(1):29 − 32. (in Chinese with English abstract)]
HU Zhenqi. Some thoughts on the high-quality development of the 14th five-year plan in the field of mine land reclamation and ecological restoration[J]. Journal of Intelligent Mine, 2021, 2（1）: 29 − 32. （in Chinese with English abstract）
[27] LUKICHEV S V,NAGOVITSIN O V. Information support of mineral mining and exploitation in the Khibiny mountains area[J]. Journal of Mining Science,2012,48(6):1025 − 1030. doi: 10.1134/S106273914806010X
[28] LIU Haoran,CHEN Shaoqing,LI Sen. Fuzzy evaluation of the impact of the open-pit limestone mining on geological environment[J]. Advanced Materials Research,2015,1073/1076:454 − 459.
[29] CHEN L C,ZHU Yukun,PAPANDREOU G,et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of the European Conference on Computer Vision (ECCV). Seattle:CVF,2018:801 − 818.
[30] BADRINARAYANAN V,KENDALL A,CIPOLLA R. SegNet:A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2481 − 2495. doi: 10.1109/TPAMI.2016.2644615
[31] LI Yehao,YAO Ting,PAN Yingwei,et al. Contextual transformer networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(2):1489 − 1500. doi: 10.1109/TPAMI.2022.3164083
[32] HU Han,ZHANG Zheng,XIE Zhenda,et al. Local relation networks for image recognition[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway:IEEE,2019:3463 − 3472.
[33] RAMACHANDRAN P,PARMAR N,VASWANI A,et al. Standalone self-attention in vision models[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems. NewYork:Curran Associates Inc,2019:68 − 80.
[34] ZHAO Hengshuang,JIA Jiaya,KOLTUN V. Exploring self-attention for image recognition[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway:IEEE,2020:10073 − 10082.
[35] ZHANG Baoyi,TANG Jiacheng,HUAN Yuke,et al. Multi-scale convolutional neural networks (CNNs) for landslide inventory mapping from remote sensing imagery and landslide susceptibility mapping (LSM)[J]. Geomatics,Natural Hazards and Risk,2024,15(1):2383309. doi: 10.1080/19475705.2024.2383309
[36] HAUT J M,FRANCO-VALIENTE J M,PAOLETTI M E,et al. Hyperspectral image analysis using cloud-based support vector machines[J]. SN Computer Science,2024,5(6):719. doi: 10.1007/s42979-024-03073-z
[37] LI Qiao,FU Xueliang,LI Honghui,et al. Advancing county-level potato cultivation area extraction:A novel approach utilizing multi-source remote sensing imagery and the Shapley additive explanations–sequential forward selection–random forest model[J]. Agriculture,2025,15(1):92. doi: 10.3390/agriculture15010092
[38] 王旭阳,张莹. 贝叶斯网络模型优化下的山地植被覆盖度遥感监测与规划技术[J]. 勘察科学技术,2024(6):51 − 55. [WANG Xuyang,ZHANG Ying. Remote sensing monitoring and planning for mountain vegetation coverage under Bayesian network model optimization[J]. Site Investigation Science and Technology,2024(6):51 − 55. (in Chinese with English abstract)]
WANG Xuyang, ZHANG Ying. Remote sensing monitoring and planning for mountain vegetation coverage under Bayesian network model optimization[J]. Site Investigation Science and Technology, 2024（6）: 51 − 55. （in Chinese with English abstract）
[39] 荣麒. 基于多头注意力机制和特征融合的新闻文本分类研究[D]. 广州:广东工业大学,2021. [RONG Qi. Research on news text classification based on multi-head attention mechanism and feature fusion[D]. Guangzhou:Guangdong University of Technology,2021. (in Chinese with English abstract)]
RONG Qi. Research on news text classification based on multi-head attention mechanism and feature fusion[D]. Guangzhou: Guangdong University of Technology, 2021. （in Chinese with English abstract）
[40] 阳珊. 小规模数据集下的建筑垃圾检测方法研究[D]. 绵阳:西南科技大学,2022. [YANG Shan. Research on object detection of construction waste on small-scale dataset[D]. Mianyang:Southwest University of Science and Technology,2022. (in Chinese with English abstract)]
YANG Shan. Research on object detection of construction waste on small-scale dataset[D]. Mianyang: Southwest University of Science and Technology, 2022. （in Chinese with English abstract）
[41] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Boston, Massachusetts, USA: Institute of Electrical and Electronics Engineers (IEEE), 2015: 1473 − 1481.
SIMONYAN K,ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Boston, Massachusetts, USA: Institute of Electrical and Electronics Engineers (IEEE),2015:1473 − 1481.
[42] LEE J,LEE I,KANG J,et al. Self-Attention graph pooling[C]//Proceedings of the 36th International Conference on Machine Learning.Long Beach, California, USA: PMLR (Proceedings of Machine Learning Research), 2019:3734-3743.
[43] LIN H T,LIN C J. A study on sigmoid kemels for SVM and the training of non-PSD kernels by SMO-type methods[J]. Submitted to Neural Computation,2003(2):1 − 32.
[44] FAN Engui. Extended tanh-fiunction method and its applications to nonlinear equations[J]. Physics Letters A,2000,277(4/5):212 − 218.
[45] REED M S. Stakeholder participation for environmental management:A literature review[J]. Biological Conservation,2008,141(10):2417 − 2431. doi: 10.1016/j.biocon.2008.07.014
[46] IOFFE S,SZEGEDY C. Batch normalization:Accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille: Proceedings of Machine Learning Research,2015:448 − 456.
[47] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation[DB/OL]. （2016-06-17）[2025-01-08]. https://arxiv.org/abs/1706.05587.
CHEN L C,PAPANDREOU G,SCHROFF F,et al. Rethinking atrous convolution for semantic image segmentation[DB/OL]. (2016-06-17)[2025-01-08]. https://arxiv.org/abs/1706.05587.
[48] LIN T Y,DOLLÁR P,GIRSHICK R,et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Seattle:CVF,2017:2117 − 2125.
[49] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[DB/OL]. （2014-11-14）[2025-01-09]. https://arxiv.org/abs/1411.4038.
LONG J,SHELHAMER E,DARRELL T. Fully convolutional networks for semantic segmentation[DB/OL]. (2014-11-14)[2025-01-09]. https://arxiv.org/abs/1411.4038.
[50] YANG Chengzhi,GUO Hongjun. A method of image semantic segmentation based on PSPNet[J]. Mathematical Problems in Engineering,2022(1):9763027
[51] HU Jie,SHEN Li,SUN Gang,et al. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Seattle:CVF,2018:7132 − 7141.
[52] SZEGEDY C,LIU Wei,JIA Yangqing,et al. Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway:IEEE,2015:1 − 9.
-
图(10)
表(5)
计量
- 文章访问数: 27
- PDF下载数: 13
- 施引文献: 0