Published 2026-02-28

Pavement cracks

Data source language:

  • English

Data reference:

  • CrackTree: Automatic crack detection from pavement images

Contributors of raw data:

  • Zou, Qin and Cao, Yu and Li, Qingquan and Mao, Qingzhou and Wang, Song

Data description:

  • These four datasets collectively serve as benchmarks for surface crack detection across road pavements and stone materials, distinguished by their varying imaging technologies and preparation methods. The CrackTree260 dataset focuses on deep learning readiness, expanding 260 visible-light, area-array pavement images into a massive training set of 35,100 samples through rigorous augmentation. In contrast, CRKWH100 and CrackLS315 prioritize high-resolution precision using line-array cameras with a 1mm ground sampling distance; the former captures 100 images under visible light, while the latter utilizes laser illumination for its 315 pavement images to enhance feature contrast. Finally, the Stone331 dataset shifts the domain to stone cutting surfaces, providing 331 visible-light area-array images accompanied by specific surface masks to ensure evaluation accuracy is constrained strictly to relevant material areas.

Keywords and subjects:

  • Pavement cracks,

Availability:

  • publicly accessible
0