"RGB-T Fissure Dataset"
Abstract
"This dataset was collected by unmanned aerial vehicle (UAV) in the ground fracture concentration area of Pingshuo Mining Area, Shuozhou City, Shanxi Province in July 2025. It consists of 1820 pairs of optical (RGB) and thermal infrared (TIR) data samples after standardized preprocessing, spatial registration and pixel-level annotation. The dataset is specifically established for performance evaluation of the RGB-T Fissure-Net network model. It can provide a reliable multi-modal benchmark for the research on ground fracture detection and remote sensing monitoring of geological disasters in mining areas. This dataset is the first mining fissure dataset that jointly utilizes optical and TIR modalities for semantic segmentation research."
Fields of Study & Subjects
Infrastructure Maintenance and Monitoring
Remote-Sensing Image Classification
Advanced Image Fusion Techniques
Publication details
- Journal / source
- IEEE DataPort
- Publication date
- 12 March 2026
- Language
- EN
- Document type
- Dataset