Circular pipe Conveyor belt line Deviation Dataset (CCDD)

Circular pipe Conveyor belt line Deviation Dataset (CCDD) is a specialized dataset designed for detecting key components of circular pipe conveyors—specifically rollers and belt lines—under realistic industrial conditions. It contains high-resolution, unaugmented images captured directly from operational sites such as coal mines, docks. The dataset includes 4,201 roller instances and 1,025 belt-line instances, with oriented bounding box (OBB) annotations. Data were collected using handheld mobile cameras following the movement trajectories of inspection robots, ensuring realistic viewpoints. Original source videos are also provided. This dataset aims to support research and practical deployment of robust visual inspection systems in complex industrial environments. If you find this dataset or project useful in your research or applications, please consider citing the following paper. This dataset is associated with the article: [BeltLineNet: A Shape-Prior-Guided Lightweight Network for Real-Time Deviation Detection in Circular Pipe Conveyors], DOI: [10.1109/JSEN.2025.3561351].