研究目的
To propose a method for automated detection and measurement of missing point regions (MPR) in mobile laser scanning (MLS) point clouds, which may cause severe road information loss.
研究成果
The proposed method effectively detects and measures MPR in MLS data, with the ε parameter significantly influencing performance. Optimal ε-value selection is recommended for accurate MPR detections. Future research should address the impact of vehicle path deviations and develop more robust road area delineation methods.
研究不足
The method relies on vehicle trajectory data, which may not accurately represent the road axis if the vehicle shifts lanes. The quality of MPR detection is influenced by the pixel size ε, requiring optimization for accurate results.
1：Experimental Design and Method Selection:
The method involves segmentation of the road area using scan-angle thresholds, rasterization of the segmented part into a binary image, identification and measurement of MPR via image processing techniques, and reparametrization of MPR parameters in relation to the vehicle path.
2：Sample Selection and Data Sources:
Two MLS datasets of urban road sections in Nanjing, China, were used for testing.
3：List of Experimental Equipment and Materials:
Mobile LiDAR system (Hi-Scan MMS from HI-TAGET) consisting of a SPAN GNSS+IMU Combined System, a single Z+F laser scanner, and five high-quality digital cameras.
4：Experimental Procedures and Operational Workflow:
The workflow includes segmentation, rasterization, detection and measurement of MPR in the image, and parameterization of MPR on the geodetic plane.
5：Data Analysis Methods:
Image processing techniques in MATLAB were used for MPR detection and measurement, including median filtering, pixel-based complement, and region properties analysis.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
