Computer Science ›› 2020, Vol. 47 ›› Issue (11): 205-211.doi: 10.11896/jsjkx.190900078

• Computer Graphics & Multimedia • Previous Articles     Next Articles

Study on Small Target Pedestrian Detection and Ranging Based on Monocular Vision

HUANG Tong-yuang, YANG Xue-jiao, XIANG Guo-hui, CHEN Liao   

  1. School of Artificial Intelligence,Chongqing University of Technology,Chongqing 401135,China
  • Received:2019-09-11 Revised:2020-02-17 Online:2020-11-15 Published:2020-11-05
  • About author:HUANG Tong-yuan,born in 1975,associate professor,postgraduate supervisor.His main research interests include machine learning,intelligent information processing,image processing and machine vision.
    YANG Xue-jiao,born in 1991,postgra-duate,is a member of China Computer Federation.Her main research interests include computer vision,target detection and monocular ranging.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China (61702063,41804112) and Postgraduate Innovation Fund Project of Chongqing University of Technology (ycx20192063).

Abstract: In order to improve the detection and ranging accuracy of long-distance pedestrians under automatic driving scenario,a pedestrian ranging algorithm is proposed based on the target detection of deep learning.Firstly,a redundant graph cutting method is proposed to detect the small target pedestrian by combining with the YOLOV3 model.And then the candidate bounds of all subgraphs are screened many times by the improved bounding box screening algorithm,and finally the pedestrian detection box is obtained.By analyzing the traditional similar triangle ranging algorithm,an improved similar triangle ranging algorithm including pitch and yaw is proposed.Finally,the transverse and longitudinal distances between pedestrian and the current vehicle are mea-sured in real time according to the pedestrian detection results.The experimental results show that,on the validation set BDD 100 K,the mAP of the proposed redundant graph cutting detection model is 6% higher than that of the original YOLOV3 model,and the mAP of small target pedestrian is improved by 3%,and has a better robustness.On the ranging test set collected by on-board camera,the combination of the redundant graph cutting method and the improved ranging algorithm improves the range measurement accuracy by 6.542% compared with the experimental results,which not only realizes long-distance range measurement,but also has higher range measurement accuracy.

Key words: Monocular ranging, Redundant graph cutting method, Similar triangles ranging, Small target detection, YOLOV3

CLC Number: 

  • TP391
[1] GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Columbus,Ohio:IEEE,2014:580-587.
[2] GIRSHICK R.Fast r-cnn[C]//Proceedings of the IEEE International Conference on Computer Vision.Santiago,Chile:ICCV,2015:1440-1448.
[3] REN S,HE K,GIRSHICK R,et al.Faster r-cnn:Towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems.Montréal,Canada:NIPS,2015:91-99.
[4] HE K,GKIOXARI G,DOLLÁR P,et al.Mask r-cnn[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice,Italy:ICCV,2017:2961-2969.
[5] LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single shotmulti-box detector[C]//EuropeanConference on Computer Vision.Cham:Springer,2016:21-37.
[6] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas,USA:IEEE,2016:779-788.
[7] REDMON J,FARHADI A.YOLO9000:better,faster,stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,Hawaii:IEEE,2017:7263-7271.
[8] REDMON J,FARHADI A.Yolov3:An incremental improve-ment [EB/OL].[2018-04-08].https://arxiv.org/abs/1804.02767.
[9] SEO D,PARK H,JO K,et al.Omnidirectional stereo visionbased vehicle detection and distance measurement for driver assistance system[C]//IECON 2013-39th Annual Conference of the IEEE Industrial Electronics Society.Vienna,Austria:IEEE,2013:5507-5511.
[10] LI X.Vehicle front obstacle measurement based on monocularvision [D].Xi'an:Xi'an University of Architecture and Technology,2013.
[11] MILKHALSKY M,SITTE J.Active motion vision for distanceestimation in autonomous mobile robots[C]//IEEE Conference on Robotics,Automation and Mechatronics.Bangkok,Thailand:IEEE,2004:1060-1065.
[12] LI T,LAI Y N,FANG T Y,et al.Research of Driving Sight Distance Detection Algorithm Based on Procurement of Special Points[J].Journal of Chongqing University of Technology(Na-tural Science),2019,33(7):37-43.
[13] YONG C W.An Improved DS Evidence Combination MethodBased on K-L Distance[J].Telecommunication Engineering,2011,5(11):27-30.
[14] WAGN Z G,ZHAO H L,WANG X.Fuzzy Ranging Method Based on Image Recognition[J].Journal of Chongqing University of Technology(Natural Science),2020,34(1):58-63.
[15] ZHOU Z,XU X B,WEN C L,et al.Optimization method forconflict evidence fusion [J].Journal of Automation,2012,38(6):976-985.
[16] WEI X.Feature fusion method based on 2DPCA and its applica-tion [J].Computer Engineering and Applications,2008,44(5):70-72.
[17] STEIN G P,MANO O,SHASHUA A.Vision-based ACC with a single camera:Bounds on range and range rate accuracy[C]//Intelligent Vehicles Symposium.Columbus,OH,USA:IEEE,2003:120-125.
[18] LIUC H,SHUAI K,YANG W R.Design of Ranging System for Embedded Vision[J].Journal of Wuhan University of Technology,2015,37(4):65-68.
[19] ZHANG J P,LOU P H,QIAN X M,et al.Research on visualguidance AGV prec-ise positioning technology for multiwindow real-time ranging [J].Journal of Instrumentation,2016,37(6):81-92.
[20] ZHONGD X,YANG Y,LIU R L,et al.Research and application of assembly robot based on monocular vision [J].Journal of Xi'an Jiaotong University,2018,52 (5):86-92.
[21] RAZA M,CHEN Z,REHMAN S U,et al.Framework for estimating distance and dimension at-tributes of pedestrians in real-time environments using monocular camera[J].Neurocomputing,2018,275:533-545.
[22] ADAMSHUK R,CARVALHO D,NEME J H Z,et al.On the applicability of inverse perspective mapping for the forward distance estimation based on the HSVcolormap[C]//2017 IEEE International Conference on Industrial Technology (ICIT).Toronto,Canada:IEEE,2017:1036-1041.
[23] ZHANG Q,ZHOU P,XUE M.A method for measuring object distance on the plane based on monocular vision[C]//Selected Papers of the Chinese Society for Optical Engineering Conferences.San Francisco,USA:SPIE,2017,10255:102554F.
[24] LIN T Y,DOLLÁR P,GIRSHICK R,et al.Feature pyramidnetworks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,Hawaii:IEEE,2017:2117-2125.
[25] NEUBECKA,GOOL L J V.Efficient Non-Maximum Suppression[C]//18th International Conference on Pattern Recognition (ICPR 2006).Hong Kong,China:IEEE Computer Society,2006:850-855.
[26] LIU Z,LU D,QIAN W,et al.A new method for increasing accuracy of distance measurement based on single visual camera[J].Optical and Quantum Electronics,2019,51(3):452-463.
[27] CAO Y,FENG Y,ZHAOL S,et al.The influence of camera attitude installation error on the accuracy of monocular vision positioning[J].Sensors and Microsystems,2012,31(12):23-26.
[28] ZHANG Z.A Flexible New Technique for Camera Calibration[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334.
[29] SILVA R A,NEME J H Z,MARGRAF E,et al.Analisis of a distance measurement system based on monocular camera[C]//XXXVIII Ibero Latin American Congress on Computational Methods in Engineering.Florianópolis,Santa Catarina,Brazil:CMEF,2017:312-326.
[1] HAO Qiang, LI Jie, ZHANG Man, WANG Lu. Spatial Non-cooperative Target Components Recognition Algorithm Based on Improved YOLOv3 [J]. Computer Science, 2022, 49(6A): 358-362.
[2] YUAN Lei, LIU Zi-yan, ZHU Ming-cheng, MA Shan-shan, CHEN Lin-zhou-ting. Improved YOLOv3 Remote Sensing Target Detection Based on Improved Dense Connection and Distributional Ranking Loss [J]. Computer Science, 2021, 48(9): 168-173.
[3] YU Han-qing, YANG Zhen, YIN Zhi-jian. Tiny YOLOv3 Target Detection Algorithm Based on Region Activation Strategy [J]. Computer Science, 2021, 48(6A): 118-121.
[4] GONG Hang, LIU Pei-shun. Detection Method of High Beam in Night Driving Vehicle [J]. Computer Science, 2021, 48(12): 256-263.
[5] LIU Yan, QIN Pin-le, ZENG Jian-chao. Multi-object Tracking Algorithm Based on YOLOv3 and Hierarchical Data Association [J]. Computer Science, 2021, 48(11A): 370-375.
[6] JIN Yu-fang, WU Xiang, DONG Hui, YU Li, ZHANG Wen-an. Improved YOLO v4 Algorithm for Safety Helmet Wearing Detection [J]. Computer Science, 2021, 48(11): 268-275.
[7] XU Shou-kun, NI Chu-han, JI Chen-chen, LI Ning. Image Caption of Safety Helmets Wearing in Construction Scene Based on YOLOv3 [J]. Computer Science, 2020, 47(8): 233-240.
[8] LIU Shu-kang, TANG Peng, JIN Wei-dong. Study on Catenary Dropper and Support Detection Based on Intelligent Data Augmentation and Improved YOLOv3 [J]. Computer Science, 2020, 47(11A): 178-182.
[9] JIAO Jiao, XIE Yong-jie, ZHANG Hua-liang and ZHANG Song. Infrared Dim and Small Target Detection Based on Background Prediction by Wavelet Filter [J]. Computer Science, 2016, 43(2): 60-63.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!