1.Hebei Provincial Key Laboratory of Earthquake Disaster Prevention and Risk Assessment, Institute of Disater Prevention, Langfang 065201, China
2.Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration, Harbin 150080, China
Objective Loess areas are highly prone to frequent earthquake-induced landslides. Rapid and accurate identification in loess regions is essential to obtain a detailed spatial distribution of loess earthquake landslides. The identification results of landslides reflect their development status and spatial distribution, serving as a theoretical foundation for research on landslide safety, vulnerability, and management. Field investigations and remote sensing technology are two primary approaches for landslide identification. However, relying solely on field surveys consumes substantial human, material, and financial resources. Similarly, depending exclusively on remote sensing technology for landslide identification does not ensure the authenticity and reliability of the database. This research applies the VGG‒Unet network for the automatic identification of loess earthquake landslides based on landslide data obtained from field surveys. Methods Firstly, based on loess earthquake landslides investigated by the earthquake landslide research team of the College of Disaster Prevention Science and Technology during field surveys in Gansu Province and Ningxia Hui Autonomous Region, the quantitative and qualitative parameters for extracting loess earthquake landslides on the satellite imagery platform through visual interpretation were systematically summarized. The quantitative parameters included stratigraphy, topography, slope, slide length, and slide width characteristics, and the qualitative parameters included planar morphology, profile, and color tone characteristics. These features supported the visual interpretation of remote sensing satellite imagery and facilitated the extraction of a database of loess landslides triggered by seismic events. Ultimately, a total of 494 original images were extracted, covering 1 052 loess landslides induced by seismic events, which served as input images to improve the model's generalization ability. Secondly, due to the exceptional feature extraction capability of the VGG16 network, its simple yet effective structural design, and its extensive pre-trained weights, it facilitated the construction of a Unet network model using it as the primary backbone feature extraction network, enhancing performance. Therefore, this study applied the VGG‒Unet network model using the Python programming language based on the PyTorch framework under the Windows 10 environment, with a GPU of NVIDIA Quadro P2000 20.9 GB, to train and validate loess earthquake landslides on the satellite imagery platform and to achieve automatic segmentation of loess earthquake landslides in the validation area, improving recognition accuracy. Three databases were established to optimize the segmentation prediction performance of the model. Database 1 contained 494 original loess earthquake landslide maps, Database 2 contained 920 loess earthquake landslide maps obtained after expansion through rotation, and Database 3 contained 613 maps obtained after expansion through cropping. The database 1 was input into the original Unet model and the VGG‒Unet model, corresponding to Experiment 1 and Experiment 2, respectively. Then, Databases 2 and 3 were input into the model that produced better results in the first two tests, namely the VGG‒Unet model, corresponding to Experiment 3 and 4, respectively. The four sets of tests were compared and analyzed to determine the optimal model performance metrics. Landslides within the study area were extracted from satellite imagery with coverage areas of 0.5~1.0 km2, ensuring that each image contained an appropriate number of landslide patches and effective training data, providing the model with sufficient landslide pixels for feature learning. A total of 494 raw images were obtained, covering 1 052 landslides, designated as Database 1. A portion of the images was expanded through rotation, yielding 920 image panels, designated as Database 2, while another portion was expanded through cropping, yielding 613 image panels, designated as Database 3. Database 1 was input into models utilizing ResNet and VGG16 as backbone features, namely the original Unet model and the VGG‒Unet model, corresponding to Experiments 1 and 2, respectively. Then, Databases 2 and 3 were input into the model that yielded superior results in the preceding two experiments, namely the VGG‒Unet model, corresponding to Experiment 3 and 4, respectively. Finally, the VGG‒Unet model, which demonstrated the best performance among the four tests, was utilized to generate prediction results for typical loess earthquake landslide areas within the validation area. The researchers compared the predicted identification results of the model with the field-investigated landslide locations one by one to ensure experimental accuracy. Results and Discussions The results showed that the VGG‒Unet model demonstrated the best training and validation performance on the dataset that was cropped and expanded to 613 frames, and performed particularly well on the unknown dataset. The accuracy was 89.57%, the MIoU was 70.13%, the F1 value was 81.11%, and the Recall reached 80.53%, and all model performance indices were high. In contrast, on the original 494-frame dataset, the performance of the original Unet model was the lowest, followed by the VGG‒Unet model, and both were lower than the results of Test 3. However, on the dataset expanded through rotation to 920 frames, although the training performance metrics of the VGG‒Unet model reached the highest values among the three tests, the validation performance metrics decreased to the lowest. This method identified loess landslides triggered by earthquakes. The model showed higher accuracy in identifying loess earthquake landslides with well-developed topographic and profile features, whereas it tended to miss landslides with ambiguous tonal and profile features, as well as small-scale landslides, and tended to misinterpret complex terrains such as uneven strata. Overall, the VGG‒Unet model effectively predicted areas prone to loess landslides. The clearer and more complete the characteristic markers of loess landslides were, the more effectively the VGG‒Unet model performed in segmentation tasks. Its high precision and accuracy enabled computers to rapidly and accurately identify target areas and determine their locations. For landslides with blurred or incomplete characteristic markers, although the model produced a limited number of misclassifications and omissions during segmentation, most areas were still accurately segmented and identified. Conclusions Therefore, the method proposed in this study can efficiently segment and recognize the same type of landslide areas and can automatically segment and recognize the locations of landslides before field investigations. This approach assists field teams in locating loess earthquake-induced landslides and provides technical support for large-scale landslide disaster investigations.
PengDa, BoJingshan, LiXiaobo,et al.Current status of research on Haiyuan earthquake-triggered landslides[J].Journal of Institute of Disaster Prevention,2017,19(3):28‒36. doi:10.3969/j.issn.1673-8047.2017.03.004
PengDa, YangShun, LiXiaobo.Distribution and mechanism of loess landslides induced by earthquake in Tongwei,Gansu Province[J].The Chinese Journal of Geological Hazard and Control,2017,28(3):31‒38. doi:10.16031/j.cnki.issn.1003-8035.2017.03.05
HuangfuWenchao, QiuHaijun, CuiPeng,et al.Quick and automatic detection of co-seismic landslides with multi-feature deep learning model[J].Scientia Sinica(Terrae),2024,54(7):2347‒2362. doi:10.1007/s11430-023-1306-8
DouHongqiang, HuangSiyi, JianWenbin,et al.Research on rapid identification technology of highway landslide in mountainous areas of southeast Fujian based on remote sensing data[J].Journal of Natural Disasters,2023,32(1):217‒227.
BaatzM, SchäpeA.Object-oriented and multi-scale image analysis in semantic networks introduction:The necessity for integration of remote sensing and GIS[C]//Proceedings of the 2nd International Symposium on Operationalization of Remote Sensing(1999).Enschede:ITC,1999:1‒7.
[12]
MarthaT R, KerleN, JettenV,et al.Characterising spectral,spatial and morphometric properties of landslides for semi-automatic detection using object-oriented methods[J].Geomorphology,2010,116(1/2):24‒36. doi:10.1016/j.geomorph.2009.10.004
[13]
LinJiayuan, WangMeimei, YangJia,et al.Landslide identification and information extraction based on optical and multispectral UAV remote sensing imagery[J].IOP Conference Series:Earth and Environmental Science,2017,57(1):012017. doi:10.1088/1755-1315/57/1/012017
[14]
DingHui, ZhangMaosheng, ZhuWeihong,et al.High resolution remote sensing for the identification of loess landslides:Example from Yan'an City[J].Northwestern Geology,2019,52(3):231‒239.
JuYuanzhen, XuQiang, JinShichao,et al.Automatic object detection of loess landslide based on deep learning[J].Geomatics and Information Science of Wuhan University,2020,45(11):1747‒1755. doi:10.13203/j.whugis20200132
JiShunping, YuDawen, ShenChaoyong,et al.Landslide detection from an open satellite imagery and digital elevation model dataset using attention boosted convolutional neural networks[J].Landslides,2020,17(6):1337‒1352. doi:10.1007/s10346-020-01353-2
[21]
YangZhaoying, HanLingyi, ZhengXiangxiang,et al.Landslide identification using remote sensing images and DEM based on convolutional neural network:A case study of loess landslide[J].Remote Sensing for Natural Resources,2022,34(2):224‒230.
JianXiaoting, ZhaoKang, ZuoXiaoqing,et al.Landslides risk identification based on Faster R-CNN—A case study in Fugong county[J].Industrial Minerals & Processing,2022,51(12):19‒24. doi:10.16283/j.cnki.hgkwyjg.2022.12.005
TanQiuyan, WuCaiyan, JiaJutao,et al.Automatic extraction of landslide from remote sensing image based on optimized FGFCM method[J].Journal of Southwest University of Science and Technology,2023,38(1):54‒60.
ChengLibo, LiJia, DuanPing,et al.A small attentional YOLO model for landslide detection from satellite remote sensing images[J].Landslides,2021,18(8):2751‒2765. doi:10.1007/s10346-021-01694-6
[28]
LiZhen, LiShanshan, GeXiaoqing.Transfer learning method for landslide extraction from GF‒1 images after the Wenchuan earthquake[J].Journal of Remote Sensing,2023,27(8):1866‒1875.
JiangWandong, XiJiangbo, LiZhenhong,et al.Landslide detection and segmentation using mask R-CNN with simulated hard samples[J].Geomatics and Information Science of Wuhan University,2023,48(12):1931‒1942. doi:10.13203/j.whugis20200692
ZhaoTong, ZhangShuangcheng, HeXiaoning,et al.Improved DeepLabV3+ model for landslide identification in high-resolution remote sensing images after earthquakes[J].National Remote Sensing Bulletin,2024,28(9):2293‒2305.
JiangSong, LiYanbo, HeXuqian,et al.Intelligent identification of landslide disaster based on deep learning of UAV images[J].China Safety Science Journal,2024,34(7):229‒238.
WangLanmin, PuXiaowu, ChenJinchang.Distribution feature and disaster risk of earthquake-induced landslide in Loess Plateau[J].City and Disaster Reduction,2019(3):33‒40. doi:10.3969/j.issn.1671-0495.2019.03.009
LinNa, FengShanshan, WangBin,et al.Rapid landslide extraction and analysis based on XGBoost from high resolution remote sensing[J].Geomatics and Information Science of Wuhan University,2024,49(8):1300‒1313. doi:10.13203/j.whugis20220296
MaoZhengjun, YuHaiyong, LiangWei,et al.Identification and feature analysis of regional loess landslides based on UAV tilt photogrammetry 3D modeling[J].Geology in China,2024,51(2):561‒576.
LiXun, YangHuan, YinZongmin,et al.Regional loess landslide recognition method research based on DEM and remote sensing image[J].Geography and Geo-Information Science,2017,33(4):86‒92. doi:10.3969/j.issn.1672-0504.2017.04.015
ZhangYuan, KongXiangsi, ZhangShuo,et al.Application of deep learning technology in remote sensing image landslide identification[J].Beijing Surveying and Mapping,2022,36(10):1385‒1390.
RonnebergerO, FischerP, BroxT.U-Net:Convolutional networks for biomedical image segmentation[EB/OL].(2015‒05‒18)[2024‒08‒15].doi:10.1007/978-3-319-24574-4_28
[49]
ChenJixing, LuPengfei, YuanZhaolin,et al.CUNet deep learning method and its application in fault recognition[J].Progress in Geophysics,2024,39(2):561‒571. doi:10.6038/pg2024HH0023
HacıefendioğluK, DemirG, BaşağaH B.Landslide detection using visualization techniques for deep convolutional neural network models[J].Natural Hazards,2021,109(1):329‒350. doi:10.1007/s11069-021-04838-y
[52]
ChenYan, LuJiahao, HuXiaochun,et al.Automated measurement method of phenotypic parameters of edible mushroom mycelium based on VGG‒UNet[J].Transactions of the Chinese Society for Agricultural Machinery,2024,55(1):233‒240.
JiangWeijie, ZhangChunju, XuBing,et al.AED-net:Semantic segmentation model for landslide recognition from remote sensing images[J].Journal of Geo-Information Science,2023,25(10):2012‒2025.
YuXianyu, YangSen.Research on automatic landslide identification based on faster R-CNN and mask R-CNN[J].Journal of Geodesy and Geodynamics,2025,45(1):1‒4. doi:10.14075/j.jgg.2024.03.090
CaiHaojie, HanHaihui, ZhangYulian,et al.Convolutional neural network landslide recognition based on terrain feature fusion[J].Journal of Earth Sciences and Environment,2022,44(3):568‒579. doi:10.19814/j.jese.2021.12016
TangXiaochuan, TuZihan, RenXuqing,et al.A multi-modal deep neural network model for forested landslide detection[J].Geomatics and Information Science of Wuhan University,2024,49(9):1566‒1573. doi:10.13203/j.whugis20230099
ChenLifu, JinYuchen, LiZhenhong,et al.SAR image earthquake landslide recognition based on multi-feature cross-fused Siamese network[J].Geomatics and Information Science of Wuhan University,2025,50(5):917‒927.
LiuJia, WuYuming, GaoXing,et al.Image recognition of co-seismic landslide based on GEE and U‒Net neural network[J].Journal of Geo-Information Science,2022,24(7):1275‒1285.
XiangXuyang, GongWenping, ZhaoFumeng,et al.Earthquake-induced landslide mapping in mountainous areas using a semantic segmentation modelcombined with a dual feature pyramid[J/OL].Journal of Earth Science,2025[2025‒06‒18].
[67]
XinLubin, HanLing, LiLiangzhi.Landslide intelligent recognition based on multi-source data fusion[J].Journal of Earth Sciences and Environment,2023,45(4):920‒928. doi:10.19814/j.jese.2022.11016