基于DSmT理论的多视角融合目标检测识别

doi:10.13973/j.cnki.robot.170507

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (1021 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
陆峰, 徐友春, 李永乐, 苏致远, 王任栋. 基于DSmT理论的多视角融合目标检测识别[J]. 机器人, 2018, 40(5): 723-733.DOI: 10.13973/j.cnki.robot.170507. LU Feng, XU Youchun, LI Yongle, SU Zhiyuan, WANG Rendong. Multi-Perspective Fusion for Object Detection and Recognition Based on DSmT. ROBOT, 2018, 40(5): 723-733. DOI: 10.13973/j.cnki.robot.170507.

摘要现有无人车在目标检测中大多依靠单一检测视角进行多传感器数据融合，受传感器检测范围的局限，难以大幅提高准确率，且对融合过程中的类别判定的高冲突情况处理较少．针对以上问题，本文基于多假设思想提出了多视角检测结果的聚类合并方法，并基于DSmT（Dezert-Samarandache theory）和时序信息，改进了冲突分配准则，降低了目标检测的漏检率与误检率．首先利用图像检测算法检测图像中的有效目标，将激光雷达的目标检测结果投影在图像平面上，通过交并比关系构建2种传感器检测结果之间的关联概率矩阵，基于多假设思想实现聚类合并，获取单帧融合检测结果．针对融合过程中可能出现的类别判定冲突情况，利用DSmT融合识别置信度，并结合时序信息对冲突重新分配，获取目标类别的准确识别结果．最后，通过实车实验对算法的有效性进行了验证．

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	陆峰
	徐友春
	李永乐
	苏致远
	王任栋

关键词 ：多传感器, 融合, 多目标, 聚类合并, DSmT, 改进冲突分配

Abstract：The existing multi-sensor fusion methods for object detection on unmanned vehicles, are mainly based on a single detection perspective. Limited by the scope of the sensor detection, it is difficult for these fusion methods to achieve high accuracy, and deal with the high conflict situation in classification when fusing the results of multiple sensors. To solve the problem and reduce the missing detection rate and false detection rate in object detection, a clustering and combination method of multi-perspective detection results is proposed based on multi-hypothesis theory, and the conflict distribution criterion is improved based on Dezert-Smarandache theory (DSmT) and time sequence information. Firstly, the image detection algorithm is applied to detecting the effective targets in the image, and the detection results of lidar are projected on the image plane. The correlation probability matrix between two kinds of detection results from different sensors is constructed by intersection-over-union. Then the cluster merging is realized based on multiple hypotheses, and the detection result based on single frame fusion is obtained. In order to handle the classification conflict situations in the fusion process, the DSmT is used to fuse the confidence degree, and the conflict is redistributed according to the time sequence information to accomplish accurate object classification. Finally, the validity of the algorithm is verified by real traffic experiment.

Key words： multi-sensor fusion multi-target clustering and combination Dezert-Samarandache theory conflict redistribution

收稿日期: 2017-08-28 录用日期: 2017-12-13

TP181

基金资助:国家自然科学基金重大项目（91220301）；国家973计划（2016YFB0100903）

通讯作者: 陆峰,1849048346@qq.com E-mail: 1849048346@qq.com

作者简介: 陆峰(1993-),男,硕士生.研究领域:智能驾驶,人工智能.
徐友春(1972-),男,博士,教授,博士生导师.研究领域:智能驾驶,人工智能.

链接本文:

https://robot.sia.cn/CN/10.13973/j.cnki.robot.170507 或 https://robot.sia.cn/CN/Y2018/V40/I5/723

[1] Chavez-Garcia R O, Aycard O. Multiple sensor fusion and class-ification for moving object detection and tracking[J]. IEEE Tr-ansactions on Intelligent Transportation Systems, 2016, 17(2):525-534.
[2] Premebida C, Carreira J, Batista J, et al. Pedestrian detection combining RGB and dense LIDAR data[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, USA:IEEE, 2014:4112-4117.
[3] Xu P, Davoine F, Denoeux T. Evidential combination of pedestrian detectors[C]//British Machine Vision Conference. London, UK:British Machine Vision Association, 2014.
[4] Jun W, Wu T, Zheng Z Y. LIDAR and vision based pedestrian detection and tracking system[C]//IEEE International Conference on Progress in Informatics and Computing. Piscataway, USA:IEEE, 2016:118-122.
[5] Girshick R. Fast R-CNN[C]//IEEE International Conference on Computer Vision. Piscataway, USA:IEEE, 2015:1440-1448.
[6] Ren S Q, He K M, Girshick R, et al. Faster R-CNN:Towardsreal-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 39(6):1137-1149.
[7] Reid D B. An algorithm for tracking multiple targets[C]//1978 IEEE Conference on Decision and Control Including the 17th Symposium on Adaptive Processes. Piscataway, USA:IEEE, 1979:1202-1211.
[8] Dollar P, Tu Z W, Perona P, et al. Integral channel features[C]//British Machine Vision Conference. London, UK:British Machine Vision Association, 2009.
[9] Felzenszwalb P F, Girshick R B, McAllester D, et al. Objectdetection with discriminatively trained part-based models[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2010, 32(9):1627-1645.
[10] Smarandache F, Dezert J. DSmT理论及其在信息融合中的应用(文集)[M].北京:国防工业出版社,2011.Smarandache F, Dezert J. Advances and applications of DSmT for information fusion (collected works)[M]. Beijing:National Defense Industry Press, 2011.
[11] 李新德,潘锦东,Dezert J.一种基于DSmT和HMM的序列飞机目标识别算法[J].自动化学报,2014,40(12):2862-2876.Li X D, Pan J D, Dezert J. A target recognition algorithm for sequential aircraft based on DSmT and HMM[J]. Acta Automatica Sinica, 2012, 40(12):2862-2876.
[12] 杨飞,朱株,龚小谨,等.基于三维激光雷达的动态障碍实时检测与跟踪[J].浙江大学学报:工学版,2012,46(9):1565-1571.Yang F, Zhu Z, Gong X J, et al. Real-time dynamic obstacle detection and tracking using 3D lidar[J]. Journal of Zhejiang University:Engineering Science, 2012, 46(9):1565-1571.