Pixel frequency based railroad surface flaw detection using active infrared thermography for Structural Health Monitoring

With rapid increase in operation and development of high-speed trains, inspection of railroad surface flaws has become an important aspect for safe and reliable operation of rail network. Non-destructive testing using active infrared thermography has been useful in determining the structural health of different structures with additional benefit of robustness in overall inspection system. This study is based on detection of artificial surface flaws on an in-service railroad. Transverse and longitudinal flaws of various dimensions were machined on rough and smooth rail surface. The railroad surface was thermally stimulated to a temperature equivalent to practical conditions. Emitted radiations from rail surface were captured by an infrared camera to detect cracks. Results show a comparison between the surface flaws on rough and smooth rail surface. Subsequently, raw infrared images were post-processed by statistical image improvement to quantitatively analyse the results. Significant change in the frequency distribution of pixel intensity is observed as the flaw size and depth changes giving a clear quantification of crack topology. A comprehensive and inexpensive solution for damage diagnosis will be offered to railway authorities for Structural Health Monitoring (SHM) and NDT by the proposed framework.

Automated summary

This study is based on the detection of artificial surface flaws on an in-service railroad, with the use of non-destructive testing using active infrared thermography. The research focuses on comparing surface flaws on rough and smooth rail surfaces, quantitatively analyzing results through statistical image improvement of raw infrared images. The proposed framework offers a comprehensive and inexpensive solution for damage diagnosis for railway authorities.