Defect classification of railway fasteners using image preprocessing and alightweight convolutional neural network

Railway fasteners are used to securely fix rails sleeper blocks. Partial wear or complete loss of these components can lead serious accidents and cause train derailments. To ensure the safety railway transportation, computer vision pattern recognition-based methods increasingly inspect infrastructure. In particular, it has become an important task detect defects in tracks. This is challenging since rail track images acquired using a measuring varying environmental conditions, at different times day poor lighting resulting often have low contrast. this study, new method proposed for classification on fasteners. The approach uses image enhancement first filter obtain high contrast image. Then, positions determined from location fastener by applying line local binary classified improved lightweight convolutional neural network (LCNN) model. Features extracted two fully connected layers developed LCNN model feature vector constructed concatenating layers. concatenated features processed number machine learning optimum classifier chosen. Experimental results show that Cubic SVM gives best with detection accuracy rate 99.7%.