Evaluation of Trans-Regional Landslide Susceptibility of Reservoir Bank Based on Transfer Component Analysis
-
摘要: 考虑到滑坡编录制作的耗时性，建立一种“可迁移”的滑坡易发性模型已越发重要.合理利用现有完整滑坡数据地区的样本集对无样本区域进行易发性预测具有重要意义.运用迁移成分分析（transfer component analysis，TCA）方法，结合深度学习卷积神经网络（convolutional neural network，CNN），尝试引入一种基于迁移学习域自适应方法的TCA-CNN模型，并以福建省两个库岸地区为例，提取11个库岸相关环境因子建立滑坡空间数据库，将有样本的池潭库区易发性模型迁移至无样本的棉花滩库区进行预测，实现跨区域滑坡易发性评价.通过对棉花滩库区进行易发性预测，结果显示：（1）采用TCA方法处理后的不同研究区数据最大均值差异（maximize mean discrepancy，MMD）明显降低（0.022），数据实现近似同分布；（2）TCA-CNN模型的跨区域预测精度为0.854，高于CNN模型（0.791），且通过历史滑坡验证其落入高、极高易发性区间的滑坡频率比占比最高（89.1%）；（3）受试者工作特性（receiver operating characteristic，ROC）曲线下面积TCA-CNN模型为0.93，高于CNN模型的0.90.可见TCA-CNN模型能够有效运用建模区的样本数据实现对无样本区域的易发性评价，且相比于传统机器模型在进行跨区域预测时具有更高、更稳定的预测准确率，具备更强的泛化能力.Abstract: It is crucial to create a "migratable" landslide susceptibility model considering the time-consuming process of recording landslides. However, a sample set of all known landslide data areas must be used in forecasting the susceptibility of unsampled regions adequately. In this paper, we attempt to develop a TCA-CNN model to enable trans-regional landslide susceptibility evaluation, which is based on the adaption domain of transfer learning by integrating the transfer component analysis (TCA) with deep learning convolutional neural network (CNN). The spatial database of landslides is constructed by extracting 11 environmental factors of the reservoir bank area, then the Mianhuatan reservoir area without samples is then predicted using the susceptibility model from the Chitan reservoir area with samples. The results show that: (1) The maximum mean discrepancy (MMD) of data from different study areas treated by TCA decreased significantly (0.022), and the data is approximately identically distributed; (2) The trans-regional prediction accuracy by the TCA-CNN model is 0.854, which is higher than that of the CNN model (0.791). And it can be verified that the proportion of landslide frequency falling into the high/extremely-high susceptibility interval is the highest (89.1%) in the historical landslide; (3) The area under the receiver operating characteristic (ROC) curve of the TCA-CNN model is 0.93, which is higher than that of the CNN model (0.90). It's obvious that the TCA-CNN model can effectively use the samples data of the modeling area to realize the susceptibility evaluation of the unsampled area. Compared with the traditional machine model, TCA-CNN model has higher and more stable prediction accuracy and stronger generalization ability in cross-region prediction.
-
图 3 研究区地理信息概况及滑坡分布
a. 福建省行政区划；b. 池潭库区‒目标域；c. 棉花滩库区‒源域
Fig. 3. Geographical information and landslide distribution in the study area
图 5 源域(池潭库区)滑坡环境因子示意
Fig. 5. Schematic diagram of landslide environmental factors in source area (Chitan reservoir area)
图 6 目标域(棉花滩库区)滑坡环境因子示意
Fig. 6. Schematic diagram of landslide environmental factors in the target area (Mianhuatan reservoir area)
图 8 不同模型下的棉花滩库区滑坡易发性区划
Fig. 8. Map of landslide susceptibility in Mianhuatan reservoir area under different models
a. TCA-CNN; b.CNN
图 9 模型跨区域预测性能同比下降率
Fig. 9. Year on year decline rate of trans-regional prediction performance of the model
表 1 研究区栅格数据来源与用途
Table 1. Source and use of grid data in the study area
数据类型 分辨率 数据来源 数据用途描述 Landsat 8 OLI_TIRS影像2021/09/02，path/row 120/412021/01/21，path/row 120/422021/02/22，path/row 120/43 全色15 m，多光谱30 m 地理空间数据云( gscloud.cn )提取归一化植被指数(normalized difference vegetation index，NDVI)；提取土地利用分类指标 DEM 30 m×30 m 地理空间数据云( gscloud.cn )提取高程、坡度、坡向、曲率等地形因子和地形湿度指数(topographic wetness index，TWI) 地质图 1∶20万 全国地质资料馆( ngac.org.cn )提取地质岩层分界线 
下载: 导出CSV
表 2 SMOTE数据扩增前后混淆矩阵
Table 2. Source domain prediction confusion matrix before and after SMOTE data amplification
数据扩增前优化的CNN模型 源域 实际 目标域 实际 非滑坡 滑坡 非滑坡 滑坡 预测 非滑坡 2 778 22 非滑坡 1 518 12 滑坡 268 12 滑坡 118 35 数据扩增后的CNN模型 源域 实际 目标域 实际 非滑坡 滑坡 非滑坡 滑坡 预测 非滑坡 2 525 275 非滑坡 984 546 滑坡 339 2 461 滑坡 95 1 435
下载: 导出CSV
表 3 各基础环境因子频率比值
Table 3. Frequency ratio of each basic environmental factor
指标因素 变量值/源域 全区栅格数/滑坡数量 频率比 变量值/目标域 全区栅格数/滑坡数量 频率比 坡度 0~9 451 017/84 1.129 0~9 54 143/38 1.397 9~16 519 186/104 1.214 9~16 96 342/48 0.992 16~23 375 979/64 1.032 16~23 73 510/40 1.083 23~33 262 212/22 0.509 23~33 58 466/22 0.749 33~75 89 097/6 0.408 33~53 22 095/5 0.450 坡向 平面 1/0 0.000 平面 78/0 0.000 北 92 861/16 1.040 北 22 146/8 0.719 东北 207 496/28 0.815 东北 41 915/12 0.570 东 230 034/33 0.866 东 38 149/25 1.304 东南 246 511/46 1.127 东南 41 190/27 1.305 南 200 569/59 1.776 南 39 604/38 1.910 西南 206 806/42 1.226 西南 33 901/25 1.468 西 211 627/16 0.456 西 30 181/9 0.594 西北 202 254/25 0.746 西北 38 121/6 0.313 北 92 163/15 0.983 北 19 271/3 0.310 曲率 ＜-1.12 34 015/5 0.896 ＜-1.12 12 988/4 0.613 -1.12~-0.31 584 005/49 0.511 -1.12~-0.31 86 848/38 0.871 -0.31~0.19 644 502/135 1.277 -0.31~0.19 131 976/62 0.935 0.19~0.79 345 859/74 1.304 0.19~0.79 60 345/42 1.385 ＞0.79 98 077/17 1.056 ＞0.79 12 399/7 1.124 TWI ＜5.69 878 092/102 0.705 ＜5.69 173 847/94 1.076 5.69~8.55 529 074/96 1.101 5.69~8.55 102 956/51 0.986 8.55~13.84 126 362/28 1.345 8.55~13.84 19 517/5 0.510 13.84~19.99 146 638/45 1.862 13.84~19.99 7 258/2 0.549 ＞19.99 18 824/9 2.901 ＞19.99 978/1 2.035 NDVI ＜0 29 332/0 0.000 ＜0 263/0 0.000 0~0.15 85 892/52 3.690 0~0.15 10 662/36 6.721 0.15~0.26 106 404/48 2.749 0.15~0.26 10 104/32 6.304 0.26~0.35 189 456/78 2.509 0.26~0.35 39 278/57 2.889 0.35~0.45 605 340/80 0.805 0.35~0.45 90 081/24 0.530 0＞0.45 690 028/22 0.194 0＞0.45 154 168/4 0.052 道路距离 ＜100 65 793/13 1.204 ＜100 37 604/65 3.441 100~200 62 426/17 1.660 100~200 21 655/21 1.930 200~300 59 939/17 1.729 200~300 16 902/21 2.473 300~400 57 655/5 0.529 300~400 14 488/20 2.748 400~500 55 801/11 1.201 400~500 12 702/8 1.254 500~600 53 582/15 1.706 500~600 11 428/1 0.174 ＞600 1 351 262/202 0.911 ＞600 189 777/17 0.178 水系距离 ＜100 62 035/57 5.603 ＜100 23 979/73 6.060 100~200 36 031/44 7.446 100~200 21 385/26 2.420 200~300 42 626/19 2.718 200~300 19 661/13 1.316 300~400 56 006/8 0.871 300~400 17 826/8 0.893 400~500 21 018/4 1.160 400~500 16 161/19 2.340 500~600 54 382/9 1.009 500~600 14 756/7 0.944 ＞600 1 435 281/139 0.591 ＞600 190 788/7 0.073 地质界线距离 ＜100 166 980/34 1.241 ＜100 33 285/13 0.777 100~200 150 919/17 0.687 100~200 30 763/16 1.035 200~300 131 706/14 0.648 200~300 26 644/11 0.822 300~400 116 448/7 0.366 300~400 23 244/11 0.942 400~500 104 562/22 1.282 400~500 20 615/2 0.193 500~600 94 819/20 1.286 500~600 17 698/11 1.237 ＞600 941 024/166 1.075 ＞600 152 307/89 1.163 滑坡涉水程度 0~0.1 833 554/180 1.317 0~0.10 225 427/50 0.442 0.1~0.3 593 353/54 0.555 0.10~0.30 38 665/32 1.647 0.3~0.52 158 988/31 1.189 0.30~0.52 19 531/40 4.077 0.52~0.82 52 736/6 0.694 0.52~0.82 9 102/10 2.187 0.82~1 69 012/9 0.795 0.82~1 11 831/21 3.533 土地分类 湖泊 20 224/0 0.000 湖泊 494/0 0.000 森林 1 434 661/141 0.599 森林 189 191/2 0.021 耕地 187 810/112 3.634 耕地 94 828/77 1.616 建设用地 31 732/1 0.192 建设用地 14 448/13 1.791 裸地 32 031/26 4.947 裸地 5 595/61 21.702 地层岩性 入侵岩 426 307/76 1.087 入侵岩 265 287/148 1.111 火山岩 242 227/21 0.529 火山岩 9 761/0 0.000 互层岩 152 118/0 0.000 变质岩 29 508/5 0.337 坚硬岩 264 425/14 0.323 变质岩 224 859/13 0.353 碎裂变质岩 229 820/61 1.619 土体 167 887/95 3.451
下载: 导出CSV
表 4 方差膨胀因子分析
Table 4. Variance expansion factor analysis
指标因素 坡度（°） 坡向（°） 曲率 TWI NDVI 道路距离（m） 水系距离（m） 地质界线距离（m） 滑坡涉水程度 土地分类 地层岩性 VIF 初始因子 4.841 4.299 45.804 3.097 29.599 2.670 2.258 2.761 1.678 6.816 2.923 剔除因子后 3.239 3.518 - 2.875 - 2.653 2.126 2.720 1.671 4.768 2.707
下载: 导出CSV
表 5 不同模型对目标域的滑坡易发性等级对比
Table 5. Comparison of landslide susceptibility grades of different models in the target area
模型 分区 栅格数 占比(%) 验证滑坡数 占比(%) 频率比 TCA-CNN 极低易发区 126 474 41.53 3 1.96 0.047 低易发区 66 503 21.84 17 11.11 0.509 中等易发区 57 407 18.85 20 13.07 0.693 高易发区 33 588 11.03 20 13.07 1.185 极高易发区 20 584 6.76 93 60.78 8.991 CNN 极低易发区 28 317 9.35 2 1.27 0.135 低易发区 40 566 13.40 4 2.53 0.189 中等易发区 47 971 15.85 12 7.59 0.479 高易发区 76 861 25.39 25 15.82 0.623 极高易发区 109 021 36.01 115 72.78 2.021
下载: 导出CSV
表 6 不同训练域模型预测指标分析
Table 6. Analysis of prediction indexes of different training domain models
模型 TCA-CNN CNN 训练指标 ACC 0.885 0.890 R2 0.647 0.634 MSE 0.088 0.092 MAE 0.192 0.231 测试指标 ACC 0.854 0.791 R2 0.571 0.413 MSE 0.107 0.147 MAE 0.182 0.240
下载: 导出CSV
表 7 不同模型跨区域预测混淆矩阵
Table 7. Trans-regional prediction confusion matrix of different models
TCA-CNN 实际 精确率(%) 非滑坡 滑坡 预测 非滑坡 1 228 302 80.26 滑坡 145 1 385 90.52 召回率(%) 89.44 82.10 85.39 CNN 实际 精确率(%) 非滑坡 滑坡 预测 非滑坡 984 546 64.31 滑坡 95 1 435 93.79 召回率(%) 91.20 72.44 79.05
下载: 导出CSV
表 8 不同模型对源域的滑坡易发性等级对比
Table 8. Comparison of landslide susceptibility grades of different models in the source area
模型 分区 栅格数 占比（%） 验证滑坡数 占比（%） 频率比 TCA-CNN 极低易发区 682 568 40.00 6 1.98 0.049 低易发区 578 724 33.91 41 14.62 0.431 中等易发区 249 546 14.62 74 26.48 1.811 高易发区 125 711 7.37 77 27.67 3.756 极高易发区 69 910 4.10 82 29.25 7.140 CNN 极低易发区 711 484 41.69 3 1.19 0.028 低易发区 498 059 29.19 34 12.25 0.420 中等易发区 213 458 12.51 70 24.90 1.991 高易发区 206 310 12.09 83 29.64 2.452 极高易发区 77 147 4.52 90 32.02 7.082
下载: 导出CSV
-
Ai, X., 2021. Construction of Earthquake Landslide Susceptibility Assessment Model Based on Machine Learning: A Case Study of Beijing Mountainous Area (Dissertation). Institute of Engineering Mechanics, China Earthquake Administration, Harbin (in Chinese with English abstract). Arabameri, A., Saha, S., Roy, J., et al., 2020. Landslide Susceptibility Evaluation and Management Using Different Machine Learning Methods in the Gallicash River Watershed, Iran. Remote Sensing, 12(3): 475. https://doi.org/10.3390/rs12030475 Chen, Z. H., 2017. Study on the Type of Bank Collapse and the Interaction between Soil and Water of Mianhuatan Reservoir in Fujian (Dissertation). Chengdu University of Technology, Chengdu (in Chinese with English abstract). Hakim, W. L., Rezaie, F., Nur, A. S., et al., 2022. Convolutional Neural Network (CNN) with Metaheuristic Optimization Algorithms for Landslide Susceptibility Mapping in Icheon, South Korea. Journal of Environmental Management, 305: 114367. https://doi.org/10.1016/j.jenvman.2021.114367 He, Q., Wang, M., Liu, K., 2021. Rapidly Assessing Earthquake-Induced Landslide Susceptibility on a Global Scale Using Random Forest. Geomorphology, 391: 107889. https://doi.org/10.1016/j.geomorph.2021.107889 Hu, Q., Zhou, Y., Wang, S. X., et al., 2019. Improving the Accuracy of Landslide Detection in "Off-Site" Area by Machine Learning Model Portability Comparison: A Case Study of Jiuzhaigou Earthquake, China. Remote Sensing, 11(21): 2530. https://doi.org/10.3390/rs11212530 Huang, F. M., 2017. Landslide Displacement Prediction and Susceptibility Assessment Based on 3S and Artificial Intelligence (Dissertation). China University of Geosciences, Wuhan (in Chinese with English abstract). Kouw, W. M., Loog, M., 2019. A Review of Single-Source Unsupervised Domain Adaptation. ArXiv, 1901.05335. http://arxiv.org/abs/1901.05335 LeCun, Y., Boser, B., Denker, J. S., et al., 1989. Backpropagation Applied to Handwritten Zip Code Recognition. Neural Computation, 1(4): 541-551. https://doi.org/10.1162/neco.1989.1.4.541 Li, S. L., Xu, Q., Tang, M. G., et al., 2020. Study on Spatial Distribution and Key Influencing Factors of Landslides in Three Gorges Reservoir Area. Earth Science, 45(1): 341-354 (in Chinese with English abstract). Li, W. B., Fan, X. M., Huang, F. M., et al., 2021. Uncertainties of Landslide Susceptibility Modeling under Different Environmental Factor Connections and Prediction Models. Earth Science, 46(10): 3777-3795 (in Chinese with English abstract). Liang, Z., 2021. Comprehensive Application and Study of Machine Learning in Susceptibility Evaluation of Shallow Landslides (Dissertation). Jilin University, Jilin (in Chinese with English abstract). Liu, R., Yang, X., Xu, C., et al., 2022. Comparative Study of Convolutional Neural Network and Conventional Machine Learning Methods for Landslide Susceptibility Mapping. Remote Sensing, 14(2): 321. https://doi.org/10.3390/rs14020321 Pan, S. J., Tsang, I. W., Kwok, J. T., et al., 2011. Domain Adaptation via Transfer Component Analysis. IEEE Transactions on Neural Networks, 22(2): 199-210. https://doi.org/10.1109/TNN.2010.2091281 Ren, R., Zhang, S. J., Sun, H. X., et al., 2021. Research on Pepper External Quality Detection Based on Transfer Learning Integrated with Convolutional Neural Network. Sensors, 21(16): 5305. https://doi.org/10.3390/s21165305 Sameen, M. I., Pradhan, B., Lee, S., 2020. Application of Convolutional Neural Networks Featuring Bayesian Optimization for Landslide Susceptibility Assessment. CATENA, 186: 104249. https://doi.org/10.1016/j.catena.2019.104249 Sun, D. L., 2019. Mapping Landslide Susceptibility Based on Machine Learning and Forecast Warning of Landslide Induced by Rainfall (Dissertation). East China Normal University, Shanghai (in Chinese with English abstract). Sun, D. L., Xu, J. H., Wen, H. J., et al., 2020. An Optimized Random Forest Model and Its Generalization Ability in Landslide Susceptibility Mapping: Application in Two Areas of Three Gorges Reservoir, China. Journal of Earth Science, 31(6): 1068-1086. https://doi.org/10.1007/s12583-020-1072-9 Tan, J. J., Yang, X. Y., Xu, Z. B., et al., 2019. Bearing Fault Diagnosis Based on Unsupervised Transfer Component Analysis and Deep Belief Network. Journal of Wuhan University of Science and Technology, 42(6): 456-462 (in Chinese with English abstract). doi: 10.3969/j.issn.1674-3644.2019.06.009 Tang, R. X., 2017. A Dissertation Submitted to China University of Geosciences for the Doctor Degree of Geological Engineering (Dissertation). China University of Geosciences, Wuhan (in Chinese with English abstract). Wang, H. J., Wang, L., Zhang, L. M., 2023. Transfer Learning Improves Landslide Susceptibility Assessment. Gondwana Research, 123: 238-254. https://doi.org/10.1016/j.gr.2022.07.008 Wang, J. J., 2015. Landslide Risk Assessment in Wanzhou County, Three Gorges (Dissertation). China University of Geosciences, Wuhan (in Chinese with English abstract). Wang, Y. M., Wu, X. L., Chen, Z. J., et al., 2019. Optimizing the Predictive Ability of Machine Learning Methods for Landslide Susceptibility Mapping Using SMOTE for Lishui City in Zhejiang Province, China. International Journal of Environmental Research and Public Health, 16(3): 368. https://doi.org/10.3390/ijerph16030368 Wu, D. R., 2011. Analyze Influencing Factors and Cure Measures of Geologic Hazard in Taining Country. Geology of Fujian, 30(3): 267-271 (in Chinese with English abstract). doi: 10.3969/j.issn.1001-3970.2011.03.012 Wu, R. Z., Hu, X. D., Mei, H. B., et al., 2021. Spatial Susceptibility Assessment of Landslides Based on Random Forest: A Case Study from Hubei Section in the Three Gorges Reservoir Area. Earth Science, 46(1): 321-330 (in Chinese with English abstract). Xu, W., Wan, Y., Zuo, T. Y., et al., 2020. Transfer Learning Based Data Feature Transfer for Fault Diagnosis. IEEE Access, 8: 76120-76129. https://doi.org/10.1109/ACCESS.2020.2989510 Yang, Z. P., Li, X. Y., Zhao, Q., et al., 2021. Key Influencing Factors Based Distribution Regularity and Deformation and Failure Response of Colluvial Landslides in Three Gorges Reservoir Area. Journal of Engineering Geology, 29(3): 617-627 (in Chinese with English abstract). Yu, X. Y., 2016. Study on The Landslide Susceptibility Evaluation Method Based on Multi-Source Data and Multi-Scale Analysis (Dissertation). China University of Geosciences, Wuhan (in Chinese with English abstract). Zhang, J., Yin, K. L., Wang, J. J., et al., 2015. Displacement Prediction of Baishuihe Landslide Based on Time Series and Pso-Svr Model. Chinese Journal of Rock Mechanics and Engineering, 34(2): 382-391 (in Chinese with English abstract). Zhang, K., Wang, J. Y., Shi, H. T., et al., 2021. A Fault Diagnosis Method Based on Improved Convolutional Neural Network for Bearings under Variable Working Conditions. Measurement, 182: 109749. https://doi.org/10.1016/j.measurement.2021.109749 Zhao, N., Zhang, X. F., Zhang, L. J., 2018. Review on Classification of Unbalanced Data. Computer Science, 45(S1): 22-27, 57 (in Chinese with English abstract). Zhao, S., Zhao, Z., 2021. A Comparative Study of Landslide Susceptibility Mapping Using SVM and PSO-SVM Models Based on Grid and Slope Units. Mathematical Problems in Engineering, 8854606. https://doi.org/10.1155/2021/8854606 Zhou, C., Yin, K. L., Cao, Y., et al., 2020. Landslide Susceptibility Assessment by Applying the Coupling Method of Radial Basis Neural Network and Adaboost: A Case Study from the Three Gorges Reservoir Area. Earth Science, 45(6): 1865-1876 (in Chinese with English abstract). Zhou, Y. C., Zhang, X. M., Wang, Y., et al., 2021. Transfer Learning and Its Application Research. Journal of Physics: Conference Series, 1920(1): 012058. https://doi.org/10.1088/1742-6596/1920/1/012058 Zhu, A. X., Miao, Y. M., Wang, R. X., et al., 2018. A Comparative Study of an Expert Knowledge-Based Model and Two Data-Driven Models for Landslide Susceptibility Mapping. CATENA, 166: 317-327. https://doi.org/10.1016/j.catena.2018.04.003 艾骁, 2021. 基于机器学习的地震滑坡易发性评估模型构建: 以北京市山区为例(博士学位论文). 哈尔滨: 中国地震局工程力学研究所. 陈中华, 2017. 福建省棉花滩水库塌岸模式及水土相互作用研究(硕士学位论文). 成都: 成都理工大学. 黄发明, 2017. 基于3S和人工智能的滑坡位移预测与易发性评价(博士学位论文). 武汉: 中国地质大学. 李松林, 许强, 汤明高, 等, 2020. 三峡库区滑坡空间发育规律及其关键影响因子. 地球科学, 45(1): 341-354. doi: 10.3799/dqkx.2017.576 李文彬, 范宣梅, 黄发明, 等, 2021. 不同环境因子联接和预测模型的滑坡易发性建模不确定性. 地球科学, 46(10): 3777-3795. doi: 10.3799/dqkx.2021.042 梁柱, 2021. 机器学习在浅层滑坡敏感性评价中的综合应用与研究(博士学位论文). 长春: 吉林大学. 孙德亮, 2019. 基于机器学习的滑坡易发性区划与降雨诱发滑坡预报预警研究(博士学位论文). 上海: 华东师范大学. 谭俊杰, 杨先勇, 徐增丙, 等, 2019. 基于无监督迁移成分分析和深度信念网络的轴承故障诊断方法. 武汉科技大学学报, 42(6): 456-462. doi: 10.3969/j.issn.1674-3644.2019.06.009 唐睿旋, 2017. 堆积层滑坡单体稳定性评估及区域易发性评价研究(博士学位论文). 武汉: 中国地质大学. 王佳佳, 2015. 三峡库区万州区滑坡灾害风险评估研究(博士学位论文). 武汉: 中国地质大学. 吴道荣, 2011. 泰宁县地质灾害影响因素及防治建议. 福建地质, 30(3): 267-271. doi: 10.3969/j.issn.1001-3970.2011.03.012 吴润泽, 胡旭东, 梅红波, 等, 2021. 基于随机森林的滑坡空间易发性评价: 以三峡库区湖北段为例. 地球科学, 46(1): 321-330. doi: 10.3799/dqkx.2020.032 杨忠平, 李绪勇, 赵茜, 等, 2021. 关键影响因子作用下三峡库区堆积层滑坡分布规律及变形破坏响应特征. 工程地质学报, 29(3): 617-627. https://www.cnki.com.cn/Article/CJFDTOTAL-GCDZ202103005.htm 于宪煜, 2016. 基于多源数据和多尺度分析的滑坡易发性评价方法研究(博士学位论文). 武汉: 中国地质大学. 张俊, 殷坤龙, 王佳佳, 等, 2015. 基于时间序列与PSO-SVR耦合模型的白水河滑坡位移预测研究. 岩石力学与工程学报, 34(2): 382-391. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201502019.htm 赵楠, 张小芳, 张利军, 2018. 不平衡数据分类研究综述. 计算机科学, 45(S1): 22-27, 57. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA2018S1005.htm 周超, 殷坤龙, 曹颖, 等, 2020. 基于集成学习与径向基神经网络耦合模型的三峡库区滑坡易发性评价. 地球科学, 45(6): 1865-1876. doi: 10.3799/dqkx.2020.071 -
点击查看大图
图(10) / 表(8)
计量
- 文章访问数: 671
- HTML全文浏览量: 141
- PDF下载量: 72
- 被引次数: 0
