Landslide Detection Using Transformers-Based Deep Learning Models
Abstract views: 6
/ 
PDF downloads: 7
Keywords:
Artificial Intelligence, Deep Learning, Image Classification, Landslide, TransformersAbstract
Landslides are caused by earthquakes and rainstorms, especially in densely populated areas.
There are different types of landslides depending on the type of material such as soil, rock, rubble, and the
type of movement such as falling, sliding, and overturning. In order to automatically detect landslides
with artificial intelligence, a dataset was first created from open source images on the internet. There are
two different classes in the dataset itself, land and landslide. Data augmentation was done in order to have
an equal number of images containing classes in the datasets obtained in different amounts. Thus, a total
of 400 image datasets were created, 200 from each class. In addition, normalized and resized operations
were performed in data preprocessing. Then the dataset is randomly divided into 80% training and 20%
validation and testing. Transformers-based deep learning models were used to perform Landslide
detection. These models are Swin transformer, Vision Transformer (ViT) and Bidirectional Encoder
representation from Image Transformers (BEiT). For detection processes, the results obtained using a
total of three models, namely ViT, BEiT, and Swin Transformer, were compared. When the obtained
results were analyzed, it was observed that both f1 score and accuracy were over 90%. Thus, in the future,
this and similar classification studies can automatically detect landslides with artificial intelligence.
Downloads
References
Highland, L. M., & Bobrowsky, P. (2008). The landslide handbook-A guide to understanding landslides (No. 1325). US Geological Survey.
Yang, J., Ding, M., Huang, W., Li, Z., Zhang, Z., Wu, J., & Peng, J. (2024). A generalized deep learning-based method for rapid co-seismic landslide mapping. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
Şener, A., & Ergen, B. (2024). LandslideSegNet: an effective deep learning network for landslide segmentation using remote sensing imagery. Earth Science Informatics, 17(5), 3963-3977.
Sreelakshmi, S., & Chandra, S. V. (2024). Visual saliency-based landslide identification using super-resolution remote sensing data. Results in Engineering, 21, 101656.
Ren, F., & Isobe, K. (2024). Versatility Evaluation of Landslide Risk with Window Sizes and Sampling Techniques Based on Deep Learning. Applied Sciences, 14(22), 10571.
Pham, T. M., Do, N., Pham, H. T., Bui, H. T., Do, T. T., & Hoang, M. V. (2024). CResU-Net: a method for landslide mapping using deep learning. Machine Learning: Science and Technology, 5(3), 035008.
Liu, D., He, M., Huang, B., Dong, Q., & Liu, S. (2025). Global landslide mapping using tibetan plateau landslide dataset and improved YOLOX. Earth Science Informatics, 18(3), 1-24.
Li, N., Feng, G., Zhao, Y., Xiong, Z., He, L., Wang, X., ... & An, Q. (2024). A deep-learning-based algorithm for landslide detection over wide areas using InSAR images considering topographic features. Sensors, 24(14), 4583.
Kaushal, A., Gupta, A. K., & Sehgal, V. K. (2024). A semantic segmentation framework with UNet-pyramid for landslide prediction using remote sensing data. Scientific Reports, 14(1), 1-23.
Kariminejad, N., Shahabi, H., Ghorbanzadeh, O., Shafaie, V., Hosseinalizadeh, M., Homayouni, S., & Pourghasemi, H. R. (2024). Evaluation of various deep learning algorithms for landslide and sinkhole detection from UAV imagery in a semi-arid environment. Earth Systems and Environment, 8(4), 1387-1398.
Huangfu, W., Qiu, H., Cui, P., Yang, D., Liu, Y., Tang, B., ... & Ullah, M. (2024). Quick and automatic detection of co-seismic landslides with multi-feature deep learning model. Science China Earth Sciences, 67(7), 2311-2325.
Ghorbanzadeh, O., Blaschke, T., Gholamnia, K., Meena, S. R., Tiede, D., & Aryal, J. (2019). Evaluation of different machine learning methods and deep-learning convolutional neural networks for landslide detection. Remote Sensing, 11(2), 196.
Gao, S., Xi, J., Li, Z., Ge, D., Guo, Z., Yu, J., ... & Xu, J. (2024). Optimal and multi-view strategic hybrid deep learning for old landslide detection in the loess plateau, Northwest China. Remote Sensing, 16(8), 1362.
Ganerød, A. J., Franch, G., Lindsay, E., & Calovi, M. (2024). Automating global landslide detection with heterogeneous ensemble deep-learning classification. Remote Sensing Applications: Society and Environment, 36, 101384.
Dang, K. B., Nguyen, C. Q., Tran, Q. C., Nguyen, H., Nguyen, T. T., Nguyen, D. A., ... & Ngo, H. H. (2024). Comparison between U-shaped structural deep learning models to detect landslide traces. Science of The Total Environment, 912, 169113.
Ghorbanzadeh, O., Shahabi, H., Piralilou, S. T., Crivellari, A., La Rosa, L. E., Atzberger, C., ... & Ghamisi, P. (2024). Contrastive self-supervised learning for globally distributed landslide detection. IEEE Access.
Cheng, G., Wang, Z., Huang, C., Yang, Y., Hu, J., Yan, X., ... & Li, H. (2024). Advances in deep learning recognition of landslides based on remote sensing images. Remote Sensing, 16(10), 1787.
Chen, T., Gao, X., Liu, G., Wang, C., Zhao, Z., Dou, J., ... & Plaza, A. J. (2024). BisDeNet: A new lightweight deep learning-based framework for efficient landslide detection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 17, 3648-3663.
Akosah, S., Gratchev, I., Kim, D. H., & Ohn, S. Y. (2024). Application of artificial intelligence and remote sensing for landslide detection and prediction: systematic review. Remote Sensing, 16(16), 2947.
Chen, X., Zhao, C., Liu, X., Zhang, S., Xi, J., & Khan, B. A. (2024). An Embedding Swin Transformer Model for Automatic Slow-moving Landslides Detection based on InSAR Products. IEEE Transactions on Geoscience and Remote Sensing.
Chandra, N., Vaidya, H., Sawant, S., & Meena, S. R. (2024). A Novel Attention-Based Generalized Efficient Layer Aggregation Network for Landslide Detection from Satellite Data in the Higher Himalayas, Nepal. Remote Sensing, 16(14), 2598.
Google Images. (n.d.). Search results for “land” and “landslide” (Creative Commons filter). Retrieved January 21, 2025, from https://images.google.com
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., ... & Guo, B. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 10012-10022).
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., ... & Houlsby, N. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929.
Bao, H., Dong, L., Piao, S., & Wei, F. (2021). Beit: Bert pre-training of image transformers. arXiv preprint arXiv:2106.08254.
Hugging Face. (n.d.). Hugging Face models. Retrieved February 1, 2025, from https://huggingface.co/models
