A FPGA-based architecture for automatic hexagonal bolts detection in railway maintenance

Rail inspection is a very important task in railway maintenance and it is periodically needed for preventing dangerous situations. Inspection is operated manually by trained human operator walking along the track searching for visual anomalies. This monitoring is unacceptable for slowness and lack of objectivity, because the results are related to the ability of the observer to recognize critical situations. The paper presents a prototypal FPGA-based architecture which automatically detects presence/absence of the fastening bolts that fix the rails to the sleepers. A simple predicting algorithm, exploiting the geometry of the railways, extracts, from the long video sequence acquired by a digital line scan camera, few windows where the presence of bolts is expected. These windows are preprocessed according to a Haar transform and then provided to a multilayer perceptron neural classifiers (MLPNCs) which reveals the presence/absence of the fastening bolts with an accuracy of 99.6% in detecting visible bolts and of 95% in detecting missing bolts. A FPGA-based architecture performs these tasks in 13.29 μs, allowing an on-the-fly analysis of a video sequence acquired up at 190 km/h.