Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Dec 11;19(24):5450.
doi: 10.3390/s19245450.

An Approach to Segment and Track-Based Pedestrian Detection from Four-Layer Laser Scanner Data

Mingfang Zhang  1 , Rui Fu  2 , Wendong Cheng  2 , Li Wang  1 , Yong Ma  2
Affiliations

An Approach to Segment and Track-Based Pedestrian Detection from Four-Layer Laser Scanner Data

Mingfang Zhang et al. Sensors (Basel). .

Abstract

Pedestrian detection is a critical perception task for autonomous driving and intelligent vehicle, and it is challenging due to the potential variation of appearance and pose of human beings as well as the partial occlusion. In this paper, we present a novel pedestrian detection method via four-layer laser scanner. The proposed approach deals with the occlusion problem by fusing the segment classification results with past knowledge integration from tracking process. First, raw point cloud is segmented into the clusters of independent objects. Then, three types of features are proposed to capture the comprehensive cues, and 18 effective features are extracted with the combination of the univariate feature selection algorithm and feature correlation analysis process. Next, based on the segment classification at individual frame, the track classification is conducted further for consecutive frames using particle filter and probability data association filter. Experimental results demonstrate that both back-propagation neural network and Adaboost classifiers based on 18 selected features have their own advantages at the segment classification stage in terms of pedestrian detection performance and computation time, and the track classification procedure can improve the detection performance particularly for partially occluded pedestrians in comparison with the single segment classification procedure.

Keywords: feature correlation analysis; laser scanner; pedestrian detection; probability data association filter; track classification.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no competing interest.

Figures

Figure 1
Figure 1
The architecture of the pedestrian detection algorithm.
Figure 2
Figure 2
IBEO LUX 4L laser scanner. (a) scan layers and vertical beam divergence [22]. (b) the laser scanner installed on the test vehicle.
Figure 3
Figure 3
The diagram of the horizontal projection curve. The dashed line denotes the fitted line of the horizontal projection points.
Figure 4
Figure 4
The labelled samples for pedestrian detection test.
Figure 5
Figure 5
The classification performance of various feature sets based on the customized threshold of the failure rate.
Figure 6
Figure 6
The tracking gate size of PDAF algorithm varies adaptively using particle filter.
Figure 7
Figure 7
The tracking algorithm scheme based on particle filter and PDAF.
Figure 8
Figure 8
ROC curves for pedestrian detection.
Figure 9
Figure 9
Pedestrian detection result in Scene 1.
Figure 9
Figure 9
Pedestrian detection result in Scene 1.
Figure 10
Figure 10
Pedestrian detection result in Scene 2.
Figure 11
Figure 11
The tracking test scenario for multiple pedestrians.
Figure 12
Figure 12
The results of tracking trajectories. (a) the tracking trajectories of multiple pedestrians in the point cloud scene. (b) the tracking trajectories in the enlarged local area A. (c) the tracking trajectories in the enlarged local area B.
Figure 13
Figure 13
The performance of the proposed track classification algorithm based on Adaboost classifier and 18 features at various ranges from the laser scanner sensor to the pedestrian.
See this image and copyright information in PMC

References

    1. Gong Z., Lin H., Zhang D., Luo Z., Zelek J., Chen Y., Wang C., Li J. A Frustum-Based Probabilistic Framework for 3D Object Detection by Fusion of LiDAR and Camera. ISPRS J. Photogramm. Remote Sens. 2020;159:90–100. doi: 10.1016/j.isprsjprs.2019.10.015. - DOI
    1. Yan P., Zhuo L., Li J., Zhang H., Zhang J. Pedestrian Attributes Recognition in Surveillance Scenarios Using Multi-Task Lightweight Convolutional Neural Network. Appl. Sci. 2019;9:4182. doi: 10.3390/app9194182. - DOI
    1. Fu X., Yu R., Zhang W., Wu J., Shao S. Delving deep into multiscale pedestrian detection via single scale feature maps. Sensors. 2018;18:1063. doi: 10.3390/s18041063. - DOI - PMC - PubMed
    1. Zhang S., Wang C., He Z., Li Q., Lin X., Zhang J., Yang C., Li J. Vehicle Global 6-DoF Pose Estimation Under Traffic Surveillance Camera. ISPRS J. Photogramm. Remote Sens. 2020;159:114–128. doi: 10.1016/j.isprsjprs.2019.11.005. - DOI
    1. Zhang M., Fu R., Morris D., Wang C. A Framework for Turning Behavior Classification at Intersections Using 3D LIDAR. IEEE Trans. Veh. Technol. 2019;68:7431–7442. doi: 10.1109/TVT.2019.2926787. - DOI

LinkOut - more resources