Rail area extraction method based on laser point cloud data

A rail area extraction method based on laser point cloud data, comprising:

• S1, collecting laser point cloud data in front of a train, and filtering out incomplete laser point cloud data caused by limitation of a field of view of a light detection and ranging (LiDAR) and laser point cloud data beyond a limited maximum detection distance from the laser point cloud data in front of the train, to thereby obtain remaining laser point cloud data;
• S2, screening out, based on a fixed distance segmentation method, target laser point cloud data from the remaining laser point cloud data, clustering the target laser point cloud data to obtain laser point cloud data clusters, and taking a main cluster of the laser point cloud data clusters to represent laser point cloud data of reference objects;
• S3, projecting the laser point cloud data of the reference objects to a horizontal plane, fitting, based on an improved differential evolution algorithm, reference curves of the reference objects, and taking a reference curve of reference objects on a left side of the train of the reference curves of the reference objects as an upper boundary, and a reference curve of reference objects on a right side of the train of the reference curves of the reference objects as a lower boundary;
• S4, selecting, based on a laser point cloud data amount-density two-step decision method, a target boundary line from the upper boundary and the lower boundary;
• S5, calculating, based on the target boundary line, a center line of a rail area;
• and
• S6, setting a threshold L, and based on a height difference between the upper boundary and the lower boundary, the threshold L, the target boundary line, and the center line of the rail area, selecting one of a rail area boundary line extension method and a rail area center line extension method segment by segment to thereby calculate the rail area.