1.中国测绘科学研究院,北京,100830
2.中国地质大学(武汉)计算机学院,湖北 武汉,430078
3.辽宁工程技术大学测绘与地理科学学院,辽宁 阜新,123000
徐胜华,博士,研究员,主要研究方向为应急地理信息服务、时空数据地学分析、个性化位置智能推荐服务。xushh@casm.ac.cn
马钰,博士生。974829885@qq.com
收稿:2024-11-20,
纸质出版:2025-06-05
移动端阅览
徐胜华,马钰,刘纪平,等.多尺度特征学习的轻量化滑坡易发性评价方法[J].武汉大学学报(信息科学版),2025,50(6):1175-1190.
XU Shenghua,MA Yu,LIU Jiping,et al.A Lightweight Network Based on Multi-scale Feature Learning for Landslide Susceptibility Assessment[J].Geomatics and Information Science of Wuhan University,2025,50(6):1175-1190.
徐胜华,马钰,刘纪平,等.多尺度特征学习的轻量化滑坡易发性评价方法[J].武汉大学学报(信息科学版),2025,50(6):1175-1190. DOI:10.13203/j.whugis20240301
XU Shenghua,MA Yu,LIU Jiping,et al.A Lightweight Network Based on Multi-scale Feature Learning for Landslide Susceptibility Assessment[J].Geomatics and Information Science of Wuhan University,2025,50(6):1175-1190. DOI:10.13203/j.whugis20240301
现有面向滑坡易发性评价的深度学习网络在不同层次的网络结构中往往仅利用单一尺度特征信息,且模型复杂造成较大的参数量和计算复杂度。提出了多尺度特征学习的轻量化滑坡易发性评价方法,利用深度特征增强模块(deep feature enhancement module,DFEM)和多尺度特征融合模块(multi-scale feature fusion module,MSFF)建立多尺度特征学习的轻量化网络,挖掘滑坡易发性与评价因子之间的深层次关联关系。引入通道增强的注意力机制和深度可分离卷积构建DFEM,动态学习不同通道特征的关注度,增强模型对重要特征的感知和利用能力;将传统卷积优化为深度可分离卷积,减少参数量和计算量,实现模型轻量化。采用空洞空间金字塔池化策略优化设计MSFF,对编码部分下采样提取的特征进行多尺度融合,在融合后的深层特征上进行信息交互,同时自适应地调整特征权重,更好地捕捉滑坡区域的全局和局部空间特征,以弥补下采样带来的特征损失。以陕西省安康市为研究区开展滑坡易发性分析,与多种经典模型对比,从模型精度和性能轻量化两个角度进行分析。结果表明,所提方法能有效降低模型训练成本,提高滑坡易发性预测的精度,易发性分区效果更好。
Objectives
2
Landslide susceptibility assessment can provide scientific basis and technical support for disaster prevention and mitigation. Additionally
factors such as engineering projects
land use
and heavy rainfall contribute to the frequent occurrence of landslides. In response to the challenges of inaccurate landslide prediction caused by using single-scale feature extraction in different levels of network structures
and the high resource demands and computational costs due to model complexity
a lightweight network based on multi-scale feature learning (MFL-LN) for landslide susceptibility assessment is proposed.
Methods
2
Combining the geological environment characteristics of landslide development
the Spearman correlation analysis was conducted on evaluation factors. This analysis led to the identification of 13 evaluation factors across four categories
including human activity factors
hydrological factors
topographical factors and vegetation factor. These factors were used to establish a landslide susceptibility evaluation factor system. The MFL-LN method employed the deep feature enhancement module (DFEM) to construct the network architecture
reducing the number of parameters and computational load to achieve model lightweighting. Additionally
it dynamically adjusted the attention to different channel features
enhancing the ability to perceive and utilize important features. Subsequently
the multi-scale feature fusion module (MSFF) was employed to integrate the features extracted through down-sampling in the encoding phase
enabling information interaction on the fused deep features and adaptively adjusting the feature weights. It can effectively capture both global and local spatial characteristics of landslides and compensate for the feature extraction loss caused by down-sampling
thereby improving the predictive capability for landslide susceptibility.
Results
2
An ablation study was conducted on the MFL-LN method
and it was compared with the five traditional models. The accuracy of the susceptibility evaluation models was analyzed from three perspectives: landslide susceptibility zoning maps
landslide hazard point density zoning statistics
and evaluation model accuracy. The hazard point density for very low and very high susceptibility areas in the MFL-LN model was 0.009 and 0.557
respectively
ensuring that the hazard point density is lower in very low susceptibility areas and highest in very high susceptibility areas. Additionally
the area under the receiver operating characteristic curve was the highest at 0.837. The proposed model was analyzed for lightweight optimization based on parameter count and floating-point operations. The parameter count of the MFL-LN model was 0.081
and the floating-point operation count was 41.102. It effectively reduced the model training costs and improved the accuracy of landslide susceptibility zoning.
Conclusions
2
By selecting Ankang City in Shaanxi Province as the study area and analyzing the historical landslide hazard data
the MFL-LN model was constructed utilizing the DFEM and MSFF with an encoder-decoder structure to evaluate landslide susceptibility. It improved computational efficiency while maintaining the performance of model
making the landslide prediction model more efficient and practical.
严俊 . 滑坡易发性预测建模的不确定性: 滑坡边界及环境因子误差的影响 [D]. 南昌 : 南昌大学 , 2022 .
YAN Jun . Uncertainties in Landslide Susceptibility Prediction Modeling: Influence of Landslide Boundary and Conditioning Factor Error [D]. Nanchang : Nanchang University , 2022 .
牛鹏飞 . 基于综合指数模型的舟曲县滑坡易发性评价 [D]. 石家庄 : 河北地质大学 , 2021 .
NIU Pengfei . Landslide Susceptibility Evaluation in Zhouqu County Based on Comprehensive Index Model [D]. Shijiazhuang : Hebei GEO University , 2021 .
ZHOU S H , ZHOU S K , TAN X . Nationwide Susceptibility Mapping of Landslides in Kenya Using the Fuzzy Analytic Hierarchy Process Model [J]. Land , 2020 , 9 ( 12 ): 535 .
崔志超 , 王俊豪 , 崔传峰 , 等 . 基于层次分析法和模糊数学相结合的甘肃东乡八丹沟泥石流易发性评价 [J]. 中国地质灾害与防治学报 , 2020 , 31 ( 1 ): 44 - 50 .
CUI Zhichao , WANG Junhao , CUI Chuanfeng , et al . Evaluation of the Susceptibility of Debris Flow in Badan Gully of Dongxiang County of Gansu Based on AHP and Fuzzy Mathematics [J]. The Chinese Journal of Geological Hazard and Control , 2020 , 31 ( 1 ): 44 - 50 .
刘福臻 , 王灵 , 肖东升 , 等 . 基于模糊综合评判法的宁南县滑坡易发性评价 [J]. 自然灾害学报 , 2021 , 30 ( 5 ): 237 - 246 .
LIU Fuzhen , WANG Ling , XIAO Dongsheng , et al . Evaluation of Landslide Susceptibility in Ningnan County Based on Fuzzy Comprehensive Evaluation [J]. Journal of Natural Disasters , 2021 , 30 ( 5 ): 237 - 246 .
胡燕 , 李德营 , 孟颂颂 , 等 . 基于证据权法的巴东县城滑坡灾害易发性评价 [J]. 地质科技通报 , 2020 , 39 ( 3 ): 187 - 194 .
HU Yan , LI Deying , MENG Songsong , et al . Landslide Susceptibility Evaluation in Badong County Based on Weights of Evidence Method [J]. Bulletin of Geological Science and Technology , 2020 , 39 ( 3 ): 187 - 194 .
周晓亭 , 黄发明 , 吴伟成 , 等 . 基于耦合信息量法选择负样本的区域滑坡易发性预测 [J]. 工程科学与技术 , 2022 , 54 ( 3 ): 25 - 35 .
ZHOU Xiaoting , HUANG Faming , WU Weicheng , et al . Regional Landslide Susceptibility Prediction Based on Negative Sample Selected by Coupling Information Value Method [J]. Advanced Engineering Sciences , 2022 , 54 ( 3 ): 25 - 35 .
WANG Q Q , GUO Y H , LI W P , et al . Predictive Modeling of Landslide Hazards in Wen County, Northwestern China Based on Information Value, Weights-of-Evidence, and Certainty Factor [J]. Geomatics, Natural Hazards and Risk , 2019 , 10 ( 1 ): 820 - 835 .
PANCHAL S , SHRIVASTAVA A K . A Comparative Study of Frequency Ratio, Shannon’s Entropy and Analytic Hierarchy Process (AHP) Models for Landslide Susceptibility Assessment [J]. ISPRS International Journal of Geo⁃Information , 2021 , 10 ( 9 ): 603 .
许嘉慧 , 张虹 , 文海家 , 等 . 基于逻辑回归的巫山县滑坡易发性区划研究 [J]. 重庆师范大学学报(自然科学版) , 2021 , 38 ( 2 ): 48 - 56 .
XU Jiahui , ZHANG Hong , WEN Haijia , et al . Landslide Susceptibility Mapping Based on Logistic Regression in Wushan County [J]. Journal of Chongqing Normal University (Natural Science) , 2021 , 38 ( 2 ): 48 - 56 .
LIU Y , ZHAO L J , BAO A M , et al . Chinese High Resolution Satellite Data and GIS-Based Assessment of Landslide Susceptibility Along Highway G30 in Guozigou Valley Using Logistic Regression and MaxEnt Model [J]. Remote Sensing , 2022 , 14 ( 15 ): 3620 .
钱紫玲 , 王平 , 李娜 , 等 . 基于信息量和逻辑回归耦合模型的黄土地震滑坡危险性分析 [J]. 地震工程学报 , 2023 , 45 ( 3 ): 706 - 715 .
QIAN Ziling , WANG Ping , LI Na , et al . Risk Analysis of Loess Seismic Landslides Based on an Information Value-Logistic Regression Coupling Model [J]. China Earthquake Engineering Journal , 2023 , 45 ( 3 ): 706 - 715 .
刘睿 , 施婌娴 , 孙德亮 , 等 . 基于GIS与随机森林的巫山县滑坡易发性区划 [J]. 重庆师范大学学报(自然科学版) , 2020 , 37 ( 3 ): 86 - 96 .
LIU Rui , SHI Shuxian , SUN Deliang , et al . Based on GIS and Random Forest Model for Landslide Susceptibility Mapping in Wushan County [J]. Journal of Chongqing Normal University (Natural Science) , 2020 , 37 ( 3 ): 86 - 96 .
吴润泽 , 胡旭东 , 梅红波 , 等 . 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 [J]. 地球科学 , 2021 , 46 ( 1 ): 321 - 330 .
WU Runze , HU Xudong , MEI Hongbo , et al . Spatial Susceptibility Assessment of Landslides Based on Random Forest: A Case Study from Hubei Section in the Three Gorges Reservoir Area [J]. Earth Science , 2021 , 46 ( 1 ): 321 - 330 .
KAVZOGLU T , TEKE A . Predictive Perfor-mances of Ensemble Machine Learning Algorithms in Landslide Susceptibility Mapping Using Random Forest, Extreme Gradient Boosting (XGBoost) and Natural Gradient Boosting (NGBoost) [J]. Arabian Journal for Science and Engineering , 2022 , 47 ( 6 ): 7367 - 7385 .
孙德亮 , 陈丹璐 , 密长林 , 等 . 基于随机森林-特征递归消除模型的可解释性缓丘岭谷地貌滑坡易发性评价 [J]. 地质力学学报 , 2023 , 29 ( 2 ): 202 - 219 .
SUN Deliang , CHEN Danlu , MI Changlin , et al . Evaluation of Landslide Susceptibility in the Gentle Hill-Valley Areas Based on the Interpretable Random Forest-Recursive Feature Elimination Model [J]. Journal of Geomechanics , 2023 , 29 ( 2 ): 202 - 219 .
FANG Z C , WANG Y , DUAN G H , et al . Landslide Susceptibility Mapping Using Rotation Forest Ensemble Technique with Different Decision Trees in the Three Gorges Reservoir Area, China [J]. Remote Sensing , 2021 , 13 ( 2 ): 238 .
ZHAO Q F , CHEN W , PENG C H , et al . Modeling Landslide Susceptibility Using an Evidential Belief Function-Based Multiclass Alternating Decision Tree and Logistic Model Tree [J]. Environmental Earth Sciences , 2022 , 81 ( 15 ): 404 .
郭天颂 , 张菊清 , 韩煜 , 等 . 基于粒子群优化支持向量机的延长县滑坡易发性评价 [J]. 地质科技情报 , 2019 , 38 ( 3 ): 236 - 243 .
GUO Tiansong , ZHANG Juqing , HAN Yu , et al . Evaluation of Landslide Susceptibility in Yanchang County Based on Particle Swarm Optimization Support Vector Machine [J]. Geological Science and Technology Information , 2019 , 38 ( 3 ): 236 - 243 .
徐胜华 , 刘纪平 , 王想红 , 等 . 熵指数融入支持向量机的滑坡灾害易发性评价方法: 以陕西省为例 [J]. 武汉大学学报(信息科学版) , 2020 , 45 ( 8 ): 1214 - 1222 .
XU Shenghua , LIU Jiping , WANG Xianghong , et al . Landslide Susceptibility Assessment Method Incorporating Index of Entropy Based on Support Vector Machine: A Case Study of Shaanxi Province [J]. Geomatics and Information Science of Wuhan University , 2020 , 45 ( 8 ): 1214 - 1222 .
林荣福 . 基于优化支持向量机模型的滑坡易发性评价: 以陕西省商洛市为例 [D]. 阜新 : 辽宁工程技术大学 , 2021 .
LIN Rongfu . Landslide Susceptibility Evaluation Based on Optimized Support Vector Machine Model - Taking Shangluo City of Shaanxi Province as an Example [D]. Fuxin : Liaoning Technical University , 2021 .
王世宝 , 庄建琦 , 郑佳 , 等 . 基于深度学习的CZ铁路康定—理塘段滑坡易发性评价 [J]. 工程地质学报 , 2022 , 30 ( 3 ): 908 - 919 .
WANG Shibao , ZHUANG Jianqi , ZHENG Jia , et al . Landslide Susceptibility Evaluation Based on Deep Learning Along Kangding-Litang Section of CZ Railway [J]. Journal of Engineering Geology , 2022 , 30 ( 3 ): 908 - 919 .
LIU R , YANG X , XU C , et al . Comparative Study of Convolutional Neural Network and Conventional Machine Learning Methods for Landslide Susceptibility Mapping [J]. Remote Sensing , 2022 , 14 ( 2 ): 321 .
YI Y N , ZHANG Z J , ZHANG W C , et al . Landslide Susceptibility Mapping Using Multiscale Sampling Strategy and Convolutional Neural Network: A Case Study in Jiuzhaigou Region [J]. CATENA , 2020 , 195 : 104851 .
WANG Z L , XU S H , LIU J P , et al . A Combination of Deep Autoencoder and Multi-scale Residual Network for Landslide Susceptibility Evaluation [J]. Remote Sensing , 2023 , 15 ( 3 ): 653 .
LIU T , CHEN T , NIU R Q , et al . Landslide Detection Mapping Employing CNN, ResNet, and DenseNet in the Three Gorges Reservoir, China [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2021 , 14 : 11417 - 11428 .
GE Y F , LIU G , TANG H M , et al . Comparative Analysis of Five Convolutional Neural Networks for Landslide Susceptibility Assessment [J]. Bulletin of Engineering Geology and the Environment , 2023 , 82 ( 10 ): 377 .
马欣蕊 , 徐胜华 , 王琢璐 , 等 . 融合自编码与密集残差网络的滑坡易发性评价 [J]. 测绘科学 , 2023 , 48 ( 7 ): 146 - 154 .
MA Xinrui , XU Shenghua , WANG Zhuolu , et al . A Combination of Autoencoder and Dense Residual Network for Landslide Susceptibility Evaluation [J ]. Science of Surveying and Mapping , 2023 , 48 ( 7 ): 146 - 154 .
蔡超 . 基于多源数据与尺度分割的滑坡易发性评价方法研究: 以上犹县为例 [D]. 赣州 : 江西理工大学 , 2021 .
CAI Chao . Landslide Susceptibility Evaluation Method Based on Multi-source Data and Scale Segmentation - The Example of Shangyou Country [D]. Ganzhou : Jiangxi University of Science and Technology , 2021 .
HOWARD A G , ZHU M L , CHEN B , et al . MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications [EB/OL]. [ 2018-04-17 ]. https://arxiv.org/abs/1704.04861 https://arxiv.org/abs/1704.04861 .
庞沛东 . 基于深度卷积神经网络的高光谱图像分类方法研究 [D]. 开封 : 河南大学 , 2021 .
PANG Peidong . Research on Hyperspectral Image Classification Method Based on Deep Convolution Neural Network [D]. Kaifeng : Henan University , 2021 .
WOO S , PARK J , LEE J Y , et al . CBAM: Convolutional Block Attention Module [C]// European Conference on Computer Vision (ECCV) , Munich, Germany , 2018 .
CHEN L C , PAPANDREOU G , SCHROFF F , et al . Rethinking Atrous Convolution for Semantic Image Segmentation [EB/OL]. [ 2023-01-17 ]. https://arxiv.org/abs/1706.05587 https://arxiv.org/abs/1706.05587 .
DAI J F , QI H Z , XIONG Y W , et al . Deformable Convolutional Networks [C]// IEEE International Conference on Computer Vision (ICCV) , Venice, Italy , 2017 .
HAN L T , LV H Y , ZHAO Y C , et al . Conv-Former: A Novel Network Combining Convolution and Self-Attention for Image Quality Assessment [J]. Sensors , 2022 , 23 ( 1 ): 427 .
AZARAFZA M , AZARAFZA M , AKGÜN H , et al . Deep Learning-Based Landslide Susceptibility Mapping [J]. Scientific Reports , 2021 , 11 ( 1 ): 24112 .
刘纪平 , 梁恩婕 , 徐胜华 , 等 . 顾及样本优化选择的多核支持向量机滑坡灾害易发性分析评价 [J]. 测绘学报 , 2022 , 51 ( 10 ): 2034 - 2045 .
LIU Jiping , LIANG Enjie , XU Shenghua , et al . Multi-kernel Support Vector Machine Considering Sample Optimization Selection for Analysis and Evaluation of Landslide Disaster Susceptibility [J]. Acta Geodaetica et Cartographica Sinica , 2022 , 51 ( 10 ): 2034 - 2045 .
ZHAO Z , LIU Z Y , XU C . Slope Unit-Based Landslide Susceptibility Mapping Using Certainty Factor, Support Vector Machine, Random Forest, CF-SVM and CF-RF Models [J]. Frontiers in Earth Science , 2021 , 9 : 589630 .
赵占骜 , 王继周 , 毛曦 , 等 . 多维CNN耦合的滑坡易发性评价方法 [J]. 武汉大学学报(信息科学版) , 2024 , 49 ( 8 ): 1466 - 1481 .
ZHAO Zhan’ao , WANG Jizhou , MAO Xi , et al . A Multi-dimensional CNN Coupled Landslide Susceptibility Assessment Method [J]. Geomatics and Information Science of Wuhan University , 2024 , 49 ( 8 ): 1466 - 1481 .
YILMAZ E O , TEKE A , KAVZOGLU T . Performance Evaluation of Depthwise Separable CNN and Random Forest Algorithms for Landslide Susceptibility Prediction [C]// IEEE International Geoscience and Remote Sensing Symposium , Kuala Lumpur, Malaysia , 2022 .
ZHAO Z A , HE Y , YAO S , et al . A Comparative Study of Different Neural Network Models for Landslide Susceptibility Mapping [J]. Advances in Space Research , 2022 , 70 ( 2 ): 383 - 401 .
JANARTHANAN S S , SUBBIAN D , SUBBARAYAN S , et al . SFCNet: Deep Learning-Based Lightweight Separable Factorized Convolution Network for Landslide Detection [J]. Journal of the Indian Society of Remote Sensing , 2023 , 51 ( 6 ): 1157 - 1170 .
0
浏览量
5
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621