[1]陶英杰,张维纬,马昕,等.面向无人机视频分析的车辆目标检测方法[J].华侨大学学报(自然科学版),2022,43(1):111-118.[doi:10.11830/ISSN.1000-5013.202011014]
　TAO Yingjie,ZHANG Weiwei,MA Xin,et al.Vehicle Target Detection Method for Unmanned Aerial Vehicle Video Analysis[J].Journal of Huaqiao University(Natural Science),2022,43(1):111-118.[doi:10.11830/ISSN.1000-5013.202011014]
点击复制

面向无人机视频分析的车辆目标检测方法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第43卷
期数:
2022年第1期
页码:
111-118
栏目:
出版日期:
2022-01-09

文章信息/Info

Title:
Vehicle Target Detection Method for Unmanned Aerial Vehicle Video Analysis
文章编号:
1000-5013(2022)01-0111-08
作者:
陶英杰1; 2;  张维纬1; 2;  马昕1; 2;  周密1; 2
1. 华侨大学 工学院, 福建 泉州 362021;2. 华侨大学 工业智能化与系统福建省高校工程研究中心, 福建 泉州 362021
Author(s):
TAO Yingjie1; 2;  ZHANG Weiwei1; 2;  MA Xin1; 2;  ZHOU Mi1; 2
1. College of Engineering, Huaqiao University, Quanzhou 362021, China; 2. Industrial Intelligence and System Fujian University Engineering Research Center, Huaqiao University, Quanzhou 362021, China
关键词:
无人机;  视频预处理;  两级过滤器;  YOLOv3;  模型压缩
Keywords:
unmanned aerial vehicle;  video preprocessing;  two-stage filter;  YOLOv3;  model compression
分类号:
TP183;TP391.41
DOI:
10.11830/ISSN.1000-5013.202011014
文献标志码:
A
摘要:
提出一种将航拍车辆视频预处理与轻量化目标检测模型结合的级联方式.首先,针对无人机拍摄的车辆视频数据大量冗余的问题,在边缘设备设置一个两级过滤器,通过帧的像素级及结构性差异过滤大量冗余帧,从而大幅减少传输到后端的检测模型的帧数;其次,针对高精度目标检测模型时延高的问题,采用通道剪枝与层剪枝结合的方法压缩YOLOv3模型并部署在PC端,实现时延和精度的均衡.实验结果表明:两级过滤器能够有效过滤90%以上的冗余帧数,相较于原模型,压缩模型在精度仅下降2%左右的情况下,检测速度提高78.3%,达到36.9帧·s-1.
Abstract:
A cascading method combining aerial vehicle video preprocessing with lightweight target detection model was proposed. Firstly, aiming at the problem of massive redundancy of vehicle video data shot by unmanned aerial vehicle, a two-stage filter was set in the edge device to filter a large number of redundant frames through the pixel level and structural difference of the frames, so as to greatly reduce the number of frames transmitted to the detection model at the back end. Secondly, to solve the problem of high delay of high-precision target detection model, a combination of channel pruning and layer pruning was used to compress the YOLOv3 model and deploy it on PC to achieve the balance of delay and precision. The experimental results show that the two-stage filter can effectively filter more than 90% of the redundant frames. Compared with the original model, the detection speed of the compression model is increased by 78.3%, reaching 36.9 frames per second, when the accuracy of the compression model is only decreased by about 2%.

参考文献/References:

[1] YANG Zi,PUN-CHENG L S C.Vehicle detection in intelligent transportation systems and its applications under varying environments: A review[J].Image and Vision Computing,2018,69:143-154.DOI:10.1016/j.imavis.2017.09.008.
[2] XIA Yingjie,WANG Chunhui,SHI Xingmin,et al.Vehicles overtaking detection using RGB-D data[J].Signal Processing,2015,112:98-109.DOI:10.1016/j.sigpro.2014.07.025.
[3] ELGAMMAL A,HARWOOD D,DAVIS L.Non-parametric model for background subtraction[C]//European Conference on Computer Vision.Dublin:Springer,2000:751-767.
[4] ZHAN Wei,JI Xiaolong.Algorithm research on moving vehicles detection[J].Procedia Engineering,2011,15:5483-5487.DOI:10.1016/j.proeng.2011.08.1017.
[5] ZHANG Wei,WU Q M J,YIN Haibing.Moving vehicles detection based on adaptive motion histogram[J].Digital Signal Processing,2010,20(3):793-805.DOI:10.1016/j.dsp.2009.10.006.
[6] JAZAYERI A,CAI Hongyuan,ZHENG Jiangyu,et al.Vehicle detection and tracking in car video based on motion model[J].IEEE Transactions on Intelligent Transportation Systems,2011,12(2):583-595.DOI:10.1109/TITS.2011.2113340.
[7] ZHU Wangjiang,HU Jie,SUN Gang,et al.A key volume mining deep framework for action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:1991-1999.DOI:10.1109/CVPR.2016.219.
[8] ZHANG Jing,LIANG Xi,WANG Meng,et al.Coarse-to-fine object detection in unmanned aerial vehicle imagery using lightweight convolutional neural network and deep motion saliency[J].Neurocomputing,2020,398:555-565.DOI:10.1016/j.neucom.2019.03.102.
[9] ZHANG Chen,CAO Qiang,JIANG Hong,et al. FFS-VA: A fast filtering system for large-scale video analytics [C]//Proceedings of the 47th International Conference on Parallel Processing.Eugene:ACM,2018:1-10.DOI:10.1145/3225058.3225103.
[10] KANG D,EMMONS J,ABUZAID F,et al.Noscope: Optimizing neural network queries over video at scale[J].Proceedings of the VLDB Endowment,2017,10(11):1586-1597.DOI:10.14778/3137628.3137664.
[11] HIMEUR Y,AIT-SAADI K,OUAMANE A.A fast and robust key-frames based video copy detection using BSIF-RMI[C]//International Conference on Signal Processing and Multimedia Applications.Vienna:IEEE Press,2016:16156288.DOI:10.5220/0005060000400047.
[12] GAO Zhen,LU Guoliang,LYU Chen,et al.Key-frame selection for automatic summarization of surveillance videos: A method of multiple change-point detection[J].Machine Vision and Applications,2018,29(7):1101-1117.DOI:10.1007/s00138-018-0954-7.
[13] WANG Haizhong,RUDY K,LI Jia,et al.Calculation of traffic flow breakdown probability to optimize link throughput[J].Applied Mathematical Modelling,2010,34:3376-3389.DOI:10.1016/j.apm.2010.02.027.
[14] HORE A,ZIOU D.Image quality metrics: PSNR vs. SSIM[C]//20th International Conference on Pattern Recognition.Istanbul:IEEE Press,2010:2366-2369.DOI:10.1109/ICPR.2010.579.
[15] WANG Zhou,BOVIK A C,SHEIKH H R,et al.Image quality assessment: From error visibility to structural similarity[J].IEEE Transactions on Image Processing,2004,13(4):600-612.DOI:10.1109/TIP.2003.819861.
[16] LIU Wei,ANGUELOV D,ERHAN D,et al.SSD: Single shot multibox detector[C]//European Conference on Computer Vision.Amsterdam:Springer,2016:21-37.DOI:10.1007/978-3-319-46448-0_2.
[17] GIRSHICK R.Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision.Santiago:IEEE Press,2015:1440-1448.DOI:10.1109/ICCV.2015.169.
[18] 姚巍巍,张洁.基于模型剪枝和半精度加速改进YOLOv3-tiny算法的实时司机违章行为检测[J].计算机系统应用,2020,29(4):41-47.DOI:10.15888/j.cnki.csa.007348.
[19] 张富凯,杨峰,李策.基于改进YOLOv3的快速车辆检测方法[J].计算机工程与应用,2019,55(2):12-20.DOI:10.3778/j.issn.1002-8331.1810-0333.
[20] 陈宏彩,任亚恒,郝存明,等.基于YOLOv3的医药玻璃瓶缺陷检测方法[J].包装工程,2020,41(7):241-246.DOI:10.19554/j.cnki.1001-3563.2020.07.034.
[21] 隋靓,党建武.基于运动目标轨迹的高速公路异常事件检测算法研究[J].计算机应用与软件,2018,35(1):246-252.DOI:10.3969/j.issn.1000-386x.2018.01.043.
[22] HAN Song,MAO Huizi,DALLY W J.Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding[C]//4th International Conference on Learning Representations.San Juan:[s.n.],2016:1-13.
[23] YAO Shuochao,ZHAO Yiran,ZHANG A,et al.Deepiot: Compressing deep neural network structures for sensing systems with a compressor-critic framework[C]//Proceedings of the 15th ACM Conference on Embedded Network Sensor Systems.Delft:ACM,2017:1-14.DOI:10.1145/3131672.3131675.
[24] ZHANG Pengyi,ZHONG Yunxin,LI Xiaoqiong.SlimYOLOv3: Narrower, faster and better for real-time UAV applications[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops.Seoul:IEEE Press,2019:37-45.DOI:10.1109/ICCVW.2019.00011.
[25] LIU Zhuang,LI Jianguo,SHEN Zhiqiang,et al.Learning efficient convolutional networks through network slimming[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice:IEEE Press,2017:2736-2744.DOI:10.1109/ICCV.2017.298.

相似文献/References:

[1]周洁,王培臣,张孟杰,等.轨迹和干扰功率联合优化的无人机主动窃听算法[J].华侨大学学报(自然科学版),2022,43(2):237.[doi:10.11830/ISSN.1000-5013.202012027]
　ZHOU Jie,WANG Peichen,ZHANG Mengjie,et al.Unmanned Aerial Vehicle Proactive Eavesdropping Algorithm Using Joint Optimization of Trajectory and Jamming Power[J].Journal of Huaqiao University(Natural Science),2022,43(1):237.[doi:10.11830/ISSN.1000-5013.202012027]

备注/Memo

备注/Memo:
收稿日期: 2020-11-04
通信作者: 张维纬(1981-),男,副教授,博士,主要从事大数据、物联网及边缘智能的研究.E-mail:178483968@qq.com.
基金项目: 国家自然科学基金面上资助项目(61976098); 福建省泉州市科技计划项目(2020C067); 华侨大学研究生科研创新基金资助项目(18014084010)http://www.hdxb.hqu.edu.cn
更新日期/Last Update: 2022-01-20