A Method for the Automatic Extraction of Support Devices in an Overhead Catenary System Based on MLS Point Clouds

A mobile laser scanning (MLS) system can acquire railway scene information quickly and provide a data foundation for regular inspections. The location of the catenary support device in an electrified has direct impact on operation power supply system. However, multi-type accounts tiny proportion whole scene, resulting its poor characteristic expression scene. Therefore, using traditional point cloud filtering or segmentation methods alone makes it difficult to achieve effective extraction device. As result, this paper proposes automatic algorithm complex devices based MLS clouds. First, obtains hierarchies pillar clouds through high stratification then realizes noise that was point-cloud-filtered Then, center is retrieved from corridor by neighborhood search, locating initial extracting are realized relatively stable spatial topological relationship between Finally, post-processing optimization method integrating filter voxelized projection designed accurate efficient feature differences other results. Furthermore, experimental part, we evaluate treatment effect six types devices, three distribution scenes, two units. results show average IoU device, unit were 97.20%, 94.29%, 96.11%, respectively. In general, proposed various different influence parameters accuracy efficiency elaborated discussion section.