Paper Key : IRJ************116
Author: Appala Naresh
Date Published: 04 Mar 2024
Abstract
Railway track fault detection is crucial for preventing accidents, especially during summer and rainy seasons. Traditional methods relying on vibrations are time-consuming. In this study, we propose a CNN-based approach to efficiently identify cracks on railway tracks using image analysis. Leveraging open-source datasets and machine learning techniques, our model significantly reduces manual inspection efforts while enhancing safety measures and cost-effectiveness. By incorporating ground truth databases and various classification methods like Random Forest and Gradient Classification, our CNN model ensures spatial and temporal coherence in defect detection, yielding accurate predictions. This innovative system streamlines fault identification, optimizing resource allocation and passenger safety in railway management