Vibration measurement is of great importance for fault diagnosis and prognosis of mechanical systems in the literature. Non-contact vibration measurement methods have been attracting growing attention in recent years. Dynamic vision is an emerging vision technology, which is developed with neuromorphic sensing principles. This paper introduces XJTU-DV, an open-source dynamic vision dataset for non-contact vibration measurement and fault diagnosis of mechanical systems. The XJTU-DV-Beam sub-dataset includes the dynamic vision data on a structural beam system, as well as the corresponding laser vibrometer data as ground truth. The XJTU-DV-Rotor and XJTU-DV-Pump sub-datasets include the dynamic vision data from a rotor and a pump test bench, respectively. The dynamic vision data are collected under different fault and operating conditions. XJTU-DV provides a data foundation for dynamic vision-based studies on vibration measurement and fault diagnosis, which may promote further development of the emerging vision algorithms for industrial applications. The dataset can be accessed at https://gr.xjtu.edu.cn/web/lixiang/xjtu-dv for detailed information. • The first open-source dynamic vision dataset for mechanical systems is constructed. • A comprehensive collection of event-based recordings under various operating conditions and mechanical types is established. • Fine-grained spatiotemporal motion data are offered to support the analysis of vibration measurement and fault diagnosis.
Li et al. (Thu,) studied this question.