"Xinjiang Regional Road Crack DataSet"
Abstract
"The dataset constructed in this paper is based on multiple field collections in Xinjiang as the core source, aiming to directly address the unique environmental challenges such as strong reflections, alternating shadows, and irregular fracture morphology in the region. Meanwhile, to enrich the diversity of crack morphology, a few characteristic samples are selected from multiple public datasets for supplementation, covering three typical types of transverse cracks, longitudinal cracks, and mesh cracks. All data were rigorously screened and de-duplicated to ensure scenario diversity and balanced category distribution."
Fields of Study & Subjects
Infrastructure Maintenance and Monitoring
Advanced Neural Network Applications
Industrial Vision Systems and Defect Detection
Publication details
- Journal / source
- IEEE DataPort
- Publication date
- 10 January 2026
- Document type
- Dataset