Research on alaser 3D profile data-driven method for rail surface anomaly detection in rail conveyors

During the long-term operation of rail conveyors, the rail surfaces are prone to wear and foreign object accumulation. If not addressed promptly, these issues can lead to unstable trolley movement or even derailment. To address the problems of delayed response and low detection accuracy in manual inspections, a track 3D profile registration deviation detection method based on the Principal Components Analysis-Normal Iterative Closest Point(PCA-NICP)algorithm is proposed, aiming to achieve high-efficiency and high-precision detection of surface anomalies in rail conveyor tracks. The method begins with data validity verification and outlier removal. Subsequently, Principal Component Analysis(PCA)and the Normal Iterative Closest Point(NICP)algorithm are employed to achieve coarse and fine registration of the profiles, respectively. Finally, the track profile is visualized, and the surface anomaly status is output. Experimental data validation shows that the detection error for track wear and foreign material thickness is less than 0.3 mm, effectively improving both detection efficiency and accuracy.