Computer Science ›› 2019, Vol. 46 ›› Issue (7): 233-237.doi: 10.11896/j.issn.1002-137X.2019.07.035

• Graphics, Image & Pattern Recognition • Previous Articles     Next Articles

Lightweight SSD Network for Real-time Object Detection in Automotive Videos

ZHANG Lin-na1,CHEN Jian-qiang1,CHEN Xiao-ling1,CEN Yi-gang2,KAN Shi-chao2   

  1. (School of Mechanical Engineering,Guizhou University,Guiyan 550025,China)1
    (School of Computer Science & Information Technology,Beijing Jiaotong University,Beijing 100044,China)2
  • Received:2018-06-18 Online:2019-07-15 Published:2019-07-15

Abstract: Vehicle and pedestrian detection are the most basic and widely studied subjectin the field of advanced driver-assistance systems (ADAS).At present,deep learning achieved the best detection performance for object detection.However,the computational cost of deep learning algorithms is very high and the algorithms often require high perfor-mance GPU.In the real applications,object detection algorithm is required to be integrated into the vehicle hardware system.So the requirement of the hardware for the algorithm can not be too high.Based on the SSD network,a lightweight SSD network was proposed for real-time objection.By resizing the input images into a smaller size and significantly reducing the node number of the fully connected layer,the network complexity could be reduced.In addition,the object detection speed was improved.A supervised training method based on the multi-stage loss function was proposed to solve the problems of image deformation and the updated parameters in the VGG low layers caused by the shrink of the input images.Furthermore,because the detection accuracy of vehicles and pedestrians would be declined after the reduction of calculations,a hierarchical image partition method was proposed to expand the training dataset,which was able to solve the object vanishing problem caused by the image shrink.Experimental results show that the proposed lightweight SSD network not only realizes real-time vehicle and pedestrian detection on a laptop,but also maintains the detection accuracy.Compared with other object detection algorithms,the optimized network achieves faster detection speed for the vehicles and pedestrians.Also,the power consuming of the laptop is reduced significantly while the detection accuracy is the same.

Key words: Advanced driver-assistance systems, Convolutional neural network, Deep learning, Object detection, SSD

CLC Number: 

  • TP391.44
[1]GIRSHICKR B,DONAHUE J,DARRELL T,et al.Region- based convolutional networks for accurate object detection and segmentation [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(1):142-158.
[2]GIRSHICKR B.Fast R-CNN[C]∥2015 IEEE International Conference on Computer Vision,ICCV 2015.Santiago,Chile,2015:1440-1448.
[3]REN S,HE K,GIRSHICKR B,et al.Faster R-CNN:towards real-time object detection with region proposal networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.
[4]LIU W,ANGUELOV D,ERHAN D,et al.SSD:single shot multibox detector[C]∥Computer Vision-ECCV 2016-14th European Conference,Amsterdam,The Netherlands,2016:21-37.
[5]REDMON J,DIVVALAS K,GIRSHICKR B,et al.You only look once:unified,real-time object detection[C]∥2016 IEEE Conference on Computer Vision and Pattern Recognition,CVPR 2016.Las Vegas,NV,USA,2016:779-788.
[6]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C]∥International Conference on Learning Representations.San Diego,USA,2015:2015-2029.
[7]CHEN S,PEI H,LAI Q,et al.Multitarget Tracking Control for Coupled Heterogeneous Inertial Agents Systems Based on Flocking Behavior[J].IEEE Transactions on Systems Man & Cybernetics Systems,2018,PP(99):1-7.
[8]DAI J,LI Y,HE K,et al.R-FCN:object detection via region-based fully convolutional networks[C]∥Advances in Neural Information Processing Systems 29:Annual Conference on Neural Information Processing Systems 2016.Barcelona,Spain,2016:379-387.
[9]REDMON J,FARHADI A.YOLO9000:better,faster,stronger[C]∥IEEE Conference on Computer Vision and Pattern Recognition,CVPR 2017.Honolulu,HI,USA,2017:6517-6525.
[10]KIM K H,HONG S,ROH B,et al.PVANET:deep but lightweight neural networks for real-time object detection[J].arXiv:1608.08021.
[11]DAI J,QI H,XIONG Y,et al.Deformable convolutional net- works[J].CoRR,abs/1703.06211,1(2),3.
[12]HUANG J,RATHOD V,SUN C,et al.Speed/accuracy trade-offs for modern convolutional object detectors[C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,Hawaii,USA,2017:3296-3297.
[13]KANG K,LI H,XIAO T,et al.Object detection in videos with tubelet proposal networks[C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,Hawaii,USA,2017:889-897.
[14]KAN S C,CEN Y G,CEN Y,et al.SURF binarization and fast codebook construction for image retrieval [J].Journal of Visual Communication & Image Representation,2017,49:104-114.
[15]GEIGER A.Are we ready for autonomous driving? The KITTI vision benchmark suite[C]∥IEEE Conference on Computer Vision and Pattern Recognition,Providence,RI,USA,2012:3354-3361.
[16]YUAN Y,YANG K,ZHANG C.Hard-aware deeply cascaded embedding[C]∥IEEE International Conference on Computer Vision,Venice,Italy,2017:814-823.
[17]EVERINGHAM M,GOOL L,WILLIAMS C K,et al.The Pascal Visual Object Classes (VOC) Challenge[J].International Journal of Computer Vision,2010,88(2):303-338.
[18]LI H,HUANG Y,ZHANG Z.An improved Faster R-CNN for same object retrieval[J].IEEE Access,2017,5:13665-13676.
[1] RAO Zhi-shuang, JIA Zhen, ZHANG Fan, LI Tian-rui. Key-Value Relational Memory Networks for Question Answering over Knowledge Graph [J]. Computer Science, 2022, 49(9): 202-207.
[2] TANG Ling-tao, WANG Di, ZHANG Lu-fei, LIU Sheng-yun. Federated Learning Scheme Based on Secure Multi-party Computation and Differential Privacy [J]. Computer Science, 2022, 49(9): 297-305.
[3] ZHOU Le-yuan, ZHANG Jian-hua, YUAN Tian-tian, CHEN Sheng-yong. Sequence-to-Sequence Chinese Continuous Sign Language Recognition and Translation with Multi- layer Attention Mechanism Fusion [J]. Computer Science, 2022, 49(9): 155-161.
[4] XU Yong-xin, ZHAO Jun-feng, WANG Ya-sha, XIE Bing, YANG Kai. Temporal Knowledge Graph Representation Learning [J]. Computer Science, 2022, 49(9): 162-171.
[5] WANG Jian, PENG Yu-qi, ZHAO Yu-fei, YANG Jian. Survey of Social Network Public Opinion Information Extraction Based on Deep Learning [J]. Computer Science, 2022, 49(8): 279-293.
[6] HAO Zhi-rong, CHEN Long, HUANG Jia-cheng. Class Discriminative Universal Adversarial Attack for Text Classification [J]. Computer Science, 2022, 49(8): 323-329.
[7] JIANG Meng-han, LI Shao-mei, ZHENG Hong-hao, ZHANG Jian-peng. Rumor Detection Model Based on Improved Position Embedding [J]. Computer Science, 2022, 49(8): 330-335.
[8] CHEN Yong-quan, JIANG Ying. Analysis Method of APP User Behavior Based on Convolutional Neural Network [J]. Computer Science, 2022, 49(8): 78-85.
[9] ZHU Cheng-zhang, HUANG Jia-er, XIAO Ya-long, WANG Han, ZOU Bei-ji. Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism [J]. Computer Science, 2022, 49(8): 113-119.
[10] LIU Dong-mei, XU Yang, WU Ze-bin, LIU Qian, SONG Bin, WEI Zhi-hui. Incremental Object Detection Method Based on Border Distance Measurement [J]. Computer Science, 2022, 49(8): 136-142.
[11] WANG Can, LIU Yong-jian, XIE Qing, MA Yan-chun. Anchor Free Object Detection Algorithm Based on Soft Label and Sample Weight Optimization [J]. Computer Science, 2022, 49(8): 157-164.
[12] SUN Qi, JI Gen-lin, ZHANG Jie. Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection [J]. Computer Science, 2022, 49(8): 172-177.
[13] HOU Yu-tao, ABULIZI Abudukelimu, ABUDUKELIMU Halidanmu. Advances in Chinese Pre-training Models [J]. Computer Science, 2022, 49(7): 148-163.
[14] ZHOU Hui, SHI Hao-chen, TU Yao-feng, HUANG Sheng-jun. Robust Deep Neural Network Learning Based on Active Sampling [J]. Computer Science, 2022, 49(7): 164-169.
[15] SU Dan-ning, CAO Gui-tao, WANG Yan-nan, WANG Hong, REN He. Survey of Deep Learning for Radar Emitter Identification Based on Small Sample [J]. Computer Science, 2022, 49(7): 226-235.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!