Object Detection on Train Bogies Using Structured Light Scanning

In this paper, a non-contact technology based on structured light scanning is introduced to inspect adhered substances on train bogie, and compute the volume of substances. The scanning system consists of a linear laser and a camera. The images recorded by the system is used to reconstruct the 3D model of a train bogie. The iterative closest point (ICP) algorithm is introduced to align the reconstructed model to an initial background model. Then a point cloud difference is obtained by subtracting the aligned point cloud from the initial model. Using the numerical integration, the volume of adhered substances is calculated from the data of the point cloud difference. The experimental results on a train bogie setup show that the structured light scanning method can reconstruct the 3D model of the bogie and accurately measure the volume of irregular adhered substances.