登录    |    注册

您好,欢迎来到中国测试科技资讯平台!

首页>《中国测试》期刊>本期导读>基于SSD-CF的无人艇目标检测跟踪方法

基于SSD-CF的无人艇目标检测跟踪方法

428    2019-02-28

¥0.00

全文售价

作者:陈欣佳1, 刘艳霞1, 洪晓斌2, 王慧芳2

作者单位:1. 华南理工大学软件学院, 广东 广州 510006;
2. 华南理工大学机械与汽车工程学院, 广东 广州 510640


关键词:无人艇;目标检测;目标跟踪;SSD;MobileNets;相关滤波


摘要:

针对目前目标检测和目标跟踪算法对无人艇运算配置要求高、速度慢等问题,该文一种提出基于SSD-CF的无人艇目标检测跟踪方法。利用MobileNets结构结合SSD目标检测算法构建轻量级卷积神经网络,实现无人艇的水面目标检测。目标检测结果作为相关滤波CF目标跟踪算法的初始输入,并在目标跟踪过程的保障其有效性。通过MODD水面船只视频数据实验表明,SSD-CF方法融合目标检测与目标跟踪算法,可有效地降低对运算力的要求,提升目标检测跟踪速度和目标位置的稳定连续性。


Unmanned boat target detection and tracking method based on SSD-CF
CHEN Xinjia1, LIU Yanxia1, HONG Xiaobin2, WANG Huifang2
1. School of Software Engineering, South China University of Technology, Guangzhou 510006, China;
2. School of Mechanical & Automotive Engineering, South China University of Technology, Guangzhou 510640, China
Abstract: Aiming at the problems of high performance and slow speed of unmanned boat operation configuration, the unmanned boat target detection and tracking method based on SSD-CF is proposed. The lightweight convolutional neural network is constructed by using MobileNets structure combined with SSD target detection algorithm to realize the surface target detection of unmanned boats. The target detection result is used as the initial input of the CF target tracking algorithm, and effectiveness of target tracking is guaranteed. The video data of MODD surface vessel shows that the SSD-CF method combines the target detection and target tracking algorithms to effectively reduce the computational force requirements and improve the target detection tracking speed and the stable continuity of the target position.
Keywords: unmanned boat;target detection;target tracking;SSD;MobileNets;CF
2019, 45(2):145-150  收稿日期: 2018-12-15;收到修改稿日期: 2019-01-03
基金项目: 广东省科技计划项目(2017B010118002);广州市科技计划项目(201802020031,201802020021)
作者简介: 陈欣佳(1993-),女,广东汕头市人,硕士研究生,专业方向为人工智能
参考文献
[1] VIOLA P, MICHAEL J. Rapid object detection using a boosted cascade of simple features[C]//Proceedings of the 2001 IEEE Computer Society Conference, 2001.
[2] DALAL N, BILL T. Histograms of oriented gradients for human detection[C]//IEEE Computer Society Conference, 2005.
[3] HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[C]//European conference on computer vision. Springer, 2014.
[4] REN S Q. Faster r-cnn:Towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2015,39(6):1137-1149.
[5] LIU W. Ssd:Single shot multibox detector[C]//European conference on computer vision, 2016.
[6] 鲁永杰, 王思明. 一种改进的运动目标跟踪方法[J]. 中国测试, 2015(7):95-98
[7] BREITENSTEIN M D. Robust tracking-by-detection using a detector confidence particle filter[C]//IEEE 12th International Conference on Computer Vision, 2009.
[8] COMANICIU D, VISVANATHAN R, PETER M. Real-time tracking of non-rigid objects using mean shift[C]//Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 2000.
[9] 朱文青, 刘艳, 卞乐,等. 基于生成式模型的目标跟踪方法综述[J]. 微处理机, 2017, 38(1):41-47
[10] BOLME D S. Visual object tracking using adaptive correlation filters[C]//2010 IEEE Conference on Computer Vision and Pattern Recognition, 2010.
[11] HENRIQUES J F. Exploiting the circulant structure of tracking-by-detection with kernels[C]//European conference on computer vision, 2012.
[12] HENRIQUES J F, CASEIRO R, MARTINS P, et al. High-speed tracking with kernelized correlation filters[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(3):583-596
[13] KRISTAN M, SULIC K V, KOVACIC S, et al. Fast image-based obstacle detection from unmanned surface vehicles[J]. IEEE Transactions on Cybernetics, 2016, 46(3):641-654.