2D image-based electrical contact surface degradation and its health assessment

Abstract Assessing the degradation of electrical contact surfaces using three-dimensional (3D) measurements is often costly and complex, limiting their use in real-time monitoring. To address this, a two-dimensional (2D) image-based approach is proposed as an efficient and practical alternative. The method uses 2D texture features, such as Gray-Level Co-occurrence Matrix (GLCM) and Local Binary Patterns (LBP), to quantify surface degradation. A novel Image-Based Degradation Index (IDI) is introduced, and its relationship with degradation extent is modeled using a Random Forest (RF) algorithm optimized by Particle Swarm Optimization (PSO). Validation against 3D data confirms the accuracy of 2D predictions. The results show that 2D imaging provides a reliable and cost-effective solution for assessing contact health, with strong agreement between 2D predictions and 3D measurements.