Pedestrian detection method based on LIDAR sensors

被引：0
|
作者：
Han X. [1 ]
Lu J. [1 ]
Li X. [1 ]
Zhao C. [1 ]
机构：
[1] School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing
来源：
Harbin Gongcheng Daxue Xuebao/Journal of Harbin Engineering University | 2019年 / 40卷 / 06期
关键词：
Autonomous driving; Environment understanding; Fast point feature histograms(FPFH); LIDAR; Object detection; Pedestrian detection; Support vector machine (SVM); Unmanned ground vehicle;
D O I：
10.11990/jheu.201803119
中图分类号：
学科分类号：
摘要：
Given that most current features based on LIDAR point clouds cannot describe the shape of pedestrian distribution, we propose a pedestrian detection method for unmanned ground vehicles. Our proposed method is based on LIDAR sensor orientation. We clustered non-ground LIDAR point clouds via DBSCAN and propose the features of the distribution of fast-point feature histograms (DFPFH) needed to train a support vector machine(SVM) classifier for pedestrian detection. The accuracy and effectiveness of the method were tested by using the KITTI OBJECT dataset and an unmanned ground vehicle. Results proved that in contrast to LIDAR-based features, DFPFH features can improve the performance of pedestrian detection effectively and can meet the real time requirement of pedestrian detection for unmanned ground vehicles. © 2019, Editorial Department of Journal of HEU. All right reserved.
引用
收藏
页码：1149 / 1154
页数：5
相关论文
共 16 条
  • [1] Dalal N., Triggs B., Histograms of oriented gradients for human detection, Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 886-893, (2005)
  • [2] Hoang V.D., Le M.H., Jo K.H., Hybrid cascade boosting machine using variant scale blocks based HOG features for pedestrian detection, Neurocomputing, 135, pp. 357-366, (2014)
  • [3] Li B., Li Y., Tian B., Et al., Part-based pedestrian detection using grammar model and ABM-HoG features, Proceedings of 2013 IEEE International Conference on Vehicular Electronics and Safety, pp. 78-83, (2013)
  • [4] Cho H., Rybski P.E., Bar-Hillel A., Et al., Real-time pedestrian detection with deformable part models, Proceedings of 2012 IEEE Intelligent Vehicles Symposium, pp. 1035-1042, (2012)
  • [5] Premebida C., Ludwig O., Nunes U., LIDAR and vision-based pedestrian detection system, Journal of Field Robotics, 26, 9, pp. 696-711, (2009)
  • [6] Haselich M., Jobgen B., Wojke N., Et al., Confidence-based pedestrian tracking in unstructured environments using 3D laser distance measurements, Proceedings of 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 4118-4123, (2014)
  • [7] Navarro-Serment L.E., Mertz C., Hebert M., Pedestrian detection and tracking using three-dimensional LADAR data, The International Journal of Robotics Research, 29, 12, pp. 1516-1528, (2010)
  • [8] Kidono K., Miyasaka T., Watanabe A., Et al., Pedestrian recognition using high-definition LIDAR, Proceedings of 2011 IEEE Intelligent Vehicles Symposium, pp. 405-410, (2011)
  • [9] Wang J., Wu T., Zheng Z., LIDAR and vision based pedestrian detection and tracking system, Proceedings of 2015 IEEE International Conference on Progress in Informatics and Computing, pp. 118-122, (2015)
  • [10] Ahtiainen J., Peynot T., Saarinen J., Et al., Learned ultra-wideband RADAR sensor model for augmented LIDAR-based traversability mapping in vegetated environments, Proceedings of Proceedings of International Conference on Information Fusion, (2015)
12