ABSTRACT
The digitisation of existing railway geometry from point clouds datasets, referred to as “twinning”, is a labourious task, which currently outweighs the perceived benefits of the resulting model. State-of-the-art methods have provided promising results, yet they cannot offer large-scale element detection required over kilometres without forfeiting precision and labour cost. The authors exploit the potential benefits of railway topology to automate the twinning process. The preliminary step involves automatically detecting masts as their positions are critical for the subsequent element detection. The method first removes vegetation and noise. Then it detects masts relative to the track centreline using the RANSAC algorithm and delivers final models in IFC format. The authors validated the method on an 18 km railway point cloud dataset and the results yielded an overall detection accuracy of 90.1% F1 score and reduced the manual twinning time by 98.6%. The proposed method lays the foundations to efficiently generate geometry-only digital twins of railway elements with no prior information.
