Vehicle target clustering identification algorithm based on 3D Lidar point cloud | |
---|---|
( Volume 3 Issue 2,February 2017 ) OPEN ACCESS | |
Author(s): | |
KONG-Dong, QU Yun-peng, KONG Peng-fei, Gao Hong-chen | |
Abstract: | |
A new vehicle target clustering identification algorithm is proposed based on the characteristics of the elevation of the structured road area and the road boundary 3D point cloud collected by the 32-line laser radar and the projection characteristics of the three-dimensional point cloud data of the vehicle target in the structured road environment. Firstly, the algorithm divides the area of interest of the smart car into six regions based on the origin of the 32-line laser radar coordinate system: right front, right rear, right, left rear, left front, left side, and the road boundary is identified based on the established structured road model, thereby reducing the interference of the obstacle outside the road boundary to the identification of the vehicle target and improving the real-time performance of the data processing. Secondly, based on the characteristics of the point cloud data of the target surface of the laser radar and the shape projection characteristics of the vehicle target, the clustering algorithm of the distance threshold is adjusted by the adaptive region. The distance threshold can be automatically based on the different regions of interest adjustment. Finally, the vehicle target is accurately identified by extracting the internal feature points of the clustering target and calculating the angle of the feature point vector or the length of the module. Experimental results show that the proposed algorithm can accurately identify the vehicle target in the structured road area, and the accuracy and robustness of the vehicle can meet the requirements of the vehicle. |
|
Paper Statistics: | Total View : 493 | Downloads : 484 | Page No: 20-23 | |
Cite this Article: | Click here to get all Styles of Citation using DOI of the article. |