116 / 2023-05-31 18:55:44

Point cloud-based algorithm for generating centrelines for different cross-sectional forms of tunnels

point cloud data; irregular alleyways; centreline extraction.

With the development of laser scanning technology, the density and accuracy of the obtained point cloud data have been greatly improved, which provides a good data base for extracting the centrelines of different forms of tunnels. The existing centreline extraction algorithms are mainly aimed at regular single tunnels. For irregular and complex tunnels, centreline extraction is more difficult and usually requires manual segmentation and then using algorithms to extract the centreline of each segment, which is labour-intensive and time-consuming. To solve this problem, this paper improves an automatic centreline extraction algorithm for tunnels.

The algorithm first projects the irregular alleyway data in the xoy plane to obtain the centreline of the projection plane E. Using the centre line of the projection plane E as a directional guide, the alleyway data is sliced to obtain the outer contour of each slice. The centroid of each slice can be calculated from the outer contour. The centrelines of all slices are then sequentially joined by a two-way point search to produce a complete centreline of the lane. The final centreline data is exported in .shp file format.

To verify the effectiveness and accuracy of the algorithm, multiple sets of irregular tunnel data of different types were used for testing. The test results show that the algorithm can better extract the data centreline of irregular alleyways and realise the automatic extraction of irregular alleyway centrelines, saving a lot of time and effort for manual segmentation. Compared with the manually extracted centrelines, the position and shape of the centrelines extracted by the algorithm are closer, indicating that it has high extraction accuracy.

Through the analysis of test results, the algorithm has strong applicability to the automatic extraction of irregular alleyway centrelines. It avoids the manual segmentation step, improves work efficiency, reduces manpower input and provides some technical support for the automatic processing and analysis of irregular tunnel data. However, the algorithm is dependent on the quality and completeness of the data, and the centreline extraction effect may be affected to a certain extent for noisy and incomplete data. Future improvements can be made in terms of improving noise immunity and adaptability to incomplete data in order to extend the application of the algorithm.

In conclusion, the algorithm provides an effective technical solution for the automatic centreline extraction of irregular tunnels. It simplifies the workflow, reduces the manual operation process, improves the work efficiency and provides a basic support for the in-depth mining and application of irregular tunnel data. With the development of navigation and positioning technology, this automated algorithm and method will have a broader application prospect.