BeltCrack: the First Sequential-image Industrial Conveyor Belt Crack Detection Dataset and Its Baseline with Triple-domain Feature Learning

TL;DR

This paper introduces BeltCrack, the first real-world industrial conveyor belt crack dataset, along with a baseline detection method utilizing triple-domain feature learning, demonstrating superior performance over existing approaches.

Contribution

The paper provides the first real-world conveyor belt crack dataset and proposes a novel triple-domain feature learning baseline for crack detection.

Findings

The dataset effectively captures real-world belt cracks.

The baseline outperforms existing detection methods.

Experimental results validate dataset usefulness and baseline effectiveness.

Abstract

Conveyor belts are important equipment in modern industry, widely applied in production and manufacturing. Their health is much critical to operational efficiency and safety. Cracks are a major threat to belt health. Currently, considering safety, how to intelligently detect belt cracks is catching an increasing attention. To implement the intelligent detection with machine learning, real crack samples are believed to be necessary. However, existing crack datasets primarily focus on pavement scenarios or synthetic data, no real-world industrial belt crack datasets at all. Cracks are a major threat to belt health. Furthermore, to validate usability and effectiveness, we propose a special baseline method with triple-domain ($i.e.$, time-space-frequency) feature hierarchical fusion learning for the two whole-new datasets. Experimental results demonstrate the availability and effectiveness of our dataset. Besides, they also show that our baseline is obviously superior to other similar detection methods. Our datasets and source codes are available at https://github.com/UESTC-nnLab/BeltCrack.