Improved Lidar Obstacle Detection Method Based on Euclidean Clustering

Chang, 刘畅 Liu; Jin, 赵津 Zhao; Zihao, 刘子豪 Liu; Xiqiao, 王玺乔 Wang; Kuncheng, 赖坤城 Lai

doi:10.3788/lop57.201105

Cited by 5 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Depth map and DBSCAN clustering algorithm represent point cloud data as depth map to cluster and segment point cloud [4][5] . The target segmentation algorithm based on the combination of Mean-shift and Euclidean clustering first carries out the rough segmentation of density clustering on the target point cloud, and then carries out the fine segmentation of the roughly segmented point cloud in combination with Euclidean clustering [6] . This method improves the detection integrity and accuracy, but Mean-Shift depends on the selection of clustering center and is easy to fall into local extreme points.…”

Section: Introductionmentioning

confidence: 99%

Point cloud target detection method combining spectral clustering and improved Euclidean clustering

Hou

Luo²,

Wang³

2023

Third International Conference on Computer Vision and Data Mining (ICCVDM 2022)

View full text Add to dashboard Cite

With the development of unmanned sensing technology, using lidar sensor to obtain 3D information of obstacles has become a research hotspot. Because the laser point cloud has the characteristics of near density and far sparse, it is easy to make the phenomenon of under segmentation of nearby objects and missing detection of distant objects in target detection, which is very easy to produce inaccurate detection results. In order to solve this problem, a combination of spectral clustering and improved European clustering algorithm is proposed, which effectively solves the problems of under segmentation and missing detection. In the experiment, the clustering effect of this algorithm is compared with the traditional European clustering algorithm and the latest point cloud detection algorithm. The real vehicle experimental results show that the average positive detection rate of the proposed algorithm is 86.9%, which is improved compared with other methods. The average time of the algorithm is 87ms, which is practical on the real vehicle platform.

show abstract

Section: Introductionmentioning

confidence: 99%