An Object Segmentation Method for Railway Surrounding Environment Based on Three-dimensional Point Cloud

Qixian Li,Yong Qin,Ping Chen,Zhipeng Wang,Xianbin Cao
DOI: https://doi.org/10.1109/sdpc49476.2020.9353151
2020-01-01
Abstract:The dramatic changes of railway surrounding environment seriously affect the railway operation safety, however the change detection mostly depends on human visual interpretation. For facilitating the change detection and reducing the human involvement, this paper proposed an automatic segmentation method for modeling railway surrounding environment using point cloud data and mainly focusing on simplifying the processing process and segmentation accuracy. Firstly, a process-based processing method is used, and only a few parameters need to be set to convert the original data to a specific model. Secondly, by analyzing the geometric and spatial features of various objects, a series of algorithms are selected to achieve the accurate segmentation of the target model. Finally, actual data along a section in Beijing-Langfang high-speed railway is used to verify the proposed method, and the results show that the method achieve a high-accuracy of segmentation.
What problem does this paper attempt to address?