Browsing by Author "Naveen Chandra"

Now showing 1 - 2 of 2

A Novel Multi-Layer Attention Boosted YOLOv10 Network for Landslide Mapping Using Remote Sensing Data
(John Wiley and Sons Inc, 2025) Naveen Chandra; Himadri Vaidya; Neelima D. Satyam; Xiaochuan Tang; Saurabh Kumar Singh; Sansar Raj Meena
Detecting landslides is a critical challenge within the remote sensing fraternity, especially given the need for timely and accurate hazard assessment. Traditional methods for identifying landslides from remote sensing data are often manual or partially automated; however, with the progress of computer vision technology, the automated methods based on deep learning algorithms have gained significant attention. Furthermore, attention mechanisms, inspired by human visual structure, have grown remarkably in various applications, including hazard studies. In this study, we leverage the capabilities of YOLO models, especially YOLOv10 and its variants, to automate the detection of landslides. We applied four prevailing attention mechanisms: CBAM, ECA, GAM, and SA. Models are trained using the Bijie landslide detection database. Moreover, the best results are unveiled based on the evaluation criteria, that is, precision, recall, f-score, and mAP. The YOLOv10m+CBAM showed the best performance with map@50-95 of 78.5%. Our results demonstrate a robust system capable of rapidly identifying and localizing landslide events with significant detection speed and accuracy improvements. This advancement augments the process of landslide detection and supports more effective disaster response and management. © 2025 The Author(s). Transactions in GIS published by John Wiley & Sons Ltd.
Optimized YOLOv8 with multi-level attention for satellite image-based landslide detection
(Elsevier Ltd, 2025) Naveen Chandra; Himadri Vaidya; Magaly Koch; Rajeshwari Bhookya; Saurabh Kumar Singh; Sansar Raj Meena
Detecting landslide events presents a significant challenge in remote sensing, especially as computer vision technologies continue to advance. Landslide detection from remotely sensed images has traditionally relied on manual or semi-automated processes. However, with the rapid development of computational resources, there has been a shift towards automatic methods leveraging deep learning algorithms. Moreover, attention models, inspired by the human visual system, have emerged substantially providing improved solutions for object detection-related problems especially, landslide mapping. Therefore, this research work introduces an enhanced YOLOv8 (nano (n), small (s), and medium (m)) network incorporating popular attention modules, specifically convolutional block attention module (CBAM), efficient channel attention (ECA), and shuffle attention (SA), for enhancing landslide detection accuracy using satellite images. The original YOLOv8 is improved by adding the attention layer discretely after the C2F module within the neck. The addition of an attention layer enables the model to concentrate on the most informative parts of the feature maps combining the capabilities of both channel and spatial attention mechanisms for detecting subtle landslide features. The experiments are conducted using the publicly available Bijie landslide detection database. Standard evaluation metrics, including precision, recall, F-score, and mean average precision (mAP), are used for quantitative analysis. Among the variants tested, YOLOv8n + ResCBAM demonstrates the most promising performance. This study underscores the model's efficacy in facilitating inventory preparation and precise landslide mapping for disaster recovery and response efforts, thereby supporting early prediction models. © 2025