Citation

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. Link to paper

Abstract

This study analyzes the potential of machine learning methods including artificial neural network (ANN), support vector machines (SVM), random forest (RF), and different deep-learning convolutional neural networks (CNNs) for landslide detection using optical data from RapidEye satellite and topographic factors. Two training zones and one test zone were used to independently evaluate the performance of different methods in the highly landslide-prone Rasuwa district in Nepal. The study tested different CNN architectures with varying window sizes and input configurations. The accuracy assessment yielded the best result of 78.26% mean Intersection over Union (mIOU) for a small window size CNN using spectral information only. The results demonstrate that deep learning CNNs can effectively identify landslides from optical satellite imagery and topographic data, with performance varying based on architecture choices and input data combinations. This research contributes to advancing automated landslide detection methods for rapid mapping in mountainous regions.

AI Visual Interpretation (Beta)