AITSAM, Muhammad, GOYAL, Gaurvi, BARTOLOZZI, Chiara and DI NUOVO, Alessandro (2025). Vibration Vision: Real-Time Machinery Fault Diagnosis with Event Cameras. In: DEL BUE, Alessio, CANTON, Cristian, PONT-TUSET, Jordi and TOMMASI, Tatiana, (eds.) Computer Vision – ECCV 2024 Workshops. Milan, Italy, September 29–October 4, 2024, Proceedings, Part XXIV. Lecture Notes in Computer Science . Cham, Springer, 293-306. [Book Section]
Documents
34496:737003
PDF
Aitsam_eccv_nevi_FaultDiagnosis.pdf - Accepted Version
Available under License Creative Commons Attribution.
Aitsam_eccv_nevi_FaultDiagnosis.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (2MB) | Preview
Abstract
Vibration-based fault detection is a critical technique in industry for identifying machinery defects and preventing operational failures. Traditional vibration monitoring methods usually rely on continuous data streams from sensors attached to the machine, which can be challenging to process in real-time. This paper introduces an innovative approach using event cameras for vibration-based fault detection. Event cameras capture changes in the scene at microsecond resolution, offering significant advantages in dynamic environments over conventional frame-based cameras due to their high temporal resolution, low energy consumption, and wide dynamic range. We propose Event-Based Frequency Mapping (EBFM) to accurately measure vibration frequencies across the observed field. By normalizing the frequency map, we create a detailed heatmap of the vibration frequencies, enabling the precise identification of anomalies and faults. The algorithm leverages the high temporal resolution of event cameras to efficiently compute dominant frequency values. Additionally, our approach includes dynamic region of interest (ROI) tracking, allowing for targeted monitoring of specific areas within the scene. Two separate experiments were conducted to test our system under different conditions. Results show that the EBFM vibration monitoring system was able to effectively measure frequencies in various lighting conditions and detect unusual machine behavior in abnormal conditions.
More Information
Statistics
Downloads
Downloads per month over past year
Metrics
Altmetric Badge
Dimensions Badge
Share
Actions (login required)
![]() |
View Item |