|Table of Contents|

Moving Target Classification and Detection AlgorithmBased on Improved YOLOv3(PDF)

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2021年04期
Page:
27-32
Research Field:
计算机科学与技术
Publishing date:

Info

Title:
Moving Target Classification and Detection AlgorithmBased on Improved YOLOv3
Author(s):
Liang Qinjia; Liu Huai; Lu Fei
School of Electrical and Automation Engineering,Nanjing Normal University,Nanjing 210023,China
Keywords:
traffic monitoring; convolutional neural networks; moving object detection
PACS:
TP391.4
DOI:
10.3969/j.issn.1672-1292.2021.04.005
Abstract:
A kind of moving target detection algorithm based on improved YOLOv3 algorithm is proposed in this paper. In order to improve the detection accuracy of YOLOv3,the boundary box regression loss function based on DIoU optimization is used. Non-maximum suppression is optimized to effectively reduce the overlap of target boxes and improve the detection accuracy. Aiming at moving target detection,a multi-center displacement detection algorithm based on target frame is proposed. The experimental results on UA-DETRAC dataset show that the detection accuracy and the fast speed can be improved by the improved algorithm. Compared with the original YOLOv3,the accuracy and recall rate are increased by 8.07% and 3.87%,respectively. The detection speed of moving target can reach 20 fps/s,which can meet the requirements of real-time detection.

References:

[1] PAN M Y,SUN J,YANG Y H,et al. Improved TQWT for marine moving target detection[J]. Journal of Systems Engineering and Electronics,2020,31(3):470-481.
[2]HU H B,XU L,ZHAO H. A spherical codebook in YUV color space for moving object detection[J]. Sensor Letters,2012,10(1-2):177-189.
[3]DU B,SUN Y J,CAI S H,et al. Object tracking in satellite videos by fusing the kernel correlation filter and the three-frame-difference algorithm[J]. IEEE Geoscience and Remote Sensing Letters,2018,15(2):168-172.
[4]WEI P C,HE F,LI J. Fast detection of moving objects based on sequential images processing[J]. Journal of Intelligent and Fuzzy Systems,2020,39(4):5037-5044.
[5]李成美,白宏阳,郭宏伟,等. 一种改进光流法的运动目标检测及跟踪算法[J]. 仪器仪表学报,2018,39(5):249-256.
[6]MANE S,MANGALE S. Moving object detection and tracking using convolutional neural networks[C]//2018 Second International Conference on Intelligent Computing and Control Systems(ICICCS). Madurai,India:IEEE,2018:1809-1813.
[7]ZOU Z X,SHI Z W,GUO Y H,et al. Object detection in 20 years:a survey[J/OL]. Computer Vision and Pattern Recognition,2019 [2021-03-08]. https://arxiv.org/abs/1905.05055.
[8]LI X,LIU Y,ZHAO Z F,et al. A deep learning approach of vehicle multitarget detection from traffic video[J/OL]. Journal of Advanced Transportation,2018(11):1-11 [2021-03-08]. https://doi.org/10.1155/2018/7075814.
[9]JU M,LUO H B,WANG Z B,et al. The application of improved YOLO V3 in multi-scale target detection[J]. Applied Sciences,2019,9(18):3775.
[10]GIRSHICK R. Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision(ICCV). Santiago,Chile:IEEE,2015:1440-1448.
[11]CHEN W P,QIAO Y T,LI Y J. Inception-SSD:An improved single shot detector for vehicle detection[J/OL]. Journal of Ambient Intelligence and Humanized Computing,2020 [2021-03-08]. https://doi.org/10.1007/S12652-020-02085W.
[12]SANG J,WU Z Y,GUO P,et al. An improved YOLOv2 for vehicle detection[J]. Sensors,2018,18(12):4272.
[13]REDMON J,FARHADI A. YOLOv3:an incremental improvement[EB/OL]. [2020-07-27]. https://arxiv.org/abs/1804.02767.
[14]REZATOFIGHI H,TSOI N,GWAK J,et al. Generalized intersection over union:a metric and a loss for bounding box regression[C]//2019 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Long Beach,CA,USA:IEEE,2019:658-666.
[15]ZHENG Z H,WANG P,LIU W,et al. Distance-IoU loss:faster and better learning for bounding box regression[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2000,34(7):12993-13000.
[16]BODLA N,SINGH B,CHELLAPPA R,et al. Soft-NMS — improving object detection with one line of code[C]//2017 IEEE International Conference on Computer Vision(ICCV). Venice,Italy:IEEE,2017:5562-5570.

Memo

Memo:
-
Last Update: 2021-12-15