基于压缩感知的变尺度目标跟踪技术

毛建森; 屈玉福

doi:10.3788/YJYXS20163105.0497

您当前的位置：

首页 >

文章列表页 >

基于压缩感知的变尺度目标跟踪技术

显示与图像 | 更新时间：2020-08-12

- 基于压缩感知的变尺度目标跟踪技术
- Tracking of variable scale object based on compressive sensing
- 液晶与显示 2016年31卷第5期页码：497-505
- 作者机构：
  
  北京航空航天大学仪器科学与光电工程学院, 北京 100191
- 作者简介：
  
  [ "毛建森(1991-),男,河南焦作人,硕士研究生,主要从事目标识别与跟踪方面的研究。E-mail:mjs110333@126.com" ]
  [ "屈玉福(1976-),男,陕西靖边人,副教授,硕士生导师,主要从事光电检测和计算成像方面的研究。E-mail:qyf@buaa.edu.cn" ]
- 基金信息：
  
  国家自然科学基金(No.51105027)()
- DOI：10.3788/YJYXS20163105.0497
  中图分类号： TP391
- 收稿日期：2015-09-12，
  
  修回日期：2015-10-09，
  
  纸质出版日期：2016-05-05
- 稿件说明：
移动端阅览
毛建森, 屈玉福,. 基于压缩感知的变尺度目标跟踪技术[J]. 液晶与显示, 2016,31(5): 497-505

MAO Jian-sen, QU Yu-fu,. Tracking of variable scale object based on compressive sensing[J]. , 2016,31(5): 497-505
毛建森, 屈玉福,. 基于压缩感知的变尺度目标跟踪技术[J]. 液晶与显示, 2016,31(5): 497-505 DOI： 10.3788/YJYXS20163105.0497.

MAO Jian-sen, QU Yu-fu,. Tracking of variable scale object based on compressive sensing[J]. , 2016,31(5): 497-505 DOI： 10.3788/YJYXS20163105.0497.

摘要

为解决变尺度目标的跟踪问题

本文基于压缩感知理论设计了一种具有目标尺度不变性的目标跟踪方法。该方法首先通过插值的方式将初始帧中要跟踪的目标扩展细化至设定的模板图像大小

提取其压缩感知变换后的低维Haar-like特征作为模板特征并初始化分类器

其次利用卡尔曼滤波对待跟踪的图像帧中目标所在位置和尺度变化趋势进行预测

然后在预测目标所在位置周围提取多个不同尺度的待测目标样本并提取其压缩感知变换后的低维Haar-like特征

最后将这些特征送入分类器进行分类得到真实目标并更新分类器。经过实验验证

本文所设计的跟踪方法的平均跟踪成功率为77%

平均中心位置误差为12像素。能够实现对运动过程中发生尺度变化的目标的有效跟踪。

Abstract

In order to solve the tracking of the variable scale object

a novel tracking algorithm with scale invariance is proposed in this paper. First

the tracked object in the initial frame is expanded to the size of the specified template image by the means of interpolation. Then the low-dimensional Haar-like features of the expanded object are extracted by compressive sensing and used as template features

and the classifier is initialized at the same time. Second

the target's center and scale variation in the frame to be processed are predicted by Kalman filtering. After that

object samples with different scales are collected around the predicted target center above and the low-dimensional Haar-like features of the object samples are extracted by compressive sensing. Finally

the low-dimensional Haar-like features are sent to the classifier to get the tracking target and the classifier is updated at the moment. Experiments show that the average success rate is 77% and the average center location error is 12 pixel. It indicates that the object tracking method proposed above is effective on the tracking of the moving target with scale variation.

关键词

Keywords

references

YILMAZ A, JAVED O, SHAH M. Object tracking:a survey[J]. ACM Computing Surveys (CSUR), 2006, 38(4):13.

侯志强,韩崇昭. 视觉跟踪技术综述[J]. 自动化学报,2006,32(4):603-617. HOU Z Q, HAN C Z. A survey of visual tracking[J]. Acta Automatica Sinica, 2006, 32(4):603-617. (in Chinese)

WU Y, LIM J, YANG M H. Online object tracking:a benchmark[C]. Proceedings of the 2013 IEEE Conference on Computer vision and pattern recognition (CVPR), Portland, OR:IEEE, 2013:2411-2418.

MEI X, LING H B. Robust visual tracking using l1 minimization[C]. Proceedings of the 2009 IEEE 12th International Conference on Computer Vision, Kyoto:IEEE, 2009:1436-1443.

SU J, YIN G S, WEI Z H, et al. Self-adaptive visual tracking method for illumination varying based on multi-feature fusion[C]. Proceedings of International Conference on Information Technology and Computer Science, 2009. ITCS 2009, Kiev:IEEE, 2009, 2:461-464.

HAN R S, JING Z L, LI Y X. Kernel based visual tracking with scale invariant features[J]. Chinese Optics Letters, 2008, 6(3):168-171.

李静宇,王延杰. 基于子空间的目标跟踪算法研究[J]. 液晶与显示, 2014,29(4):617-622. LI J Y, WANG Y J. Subspace based target tracking algorithm[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(4):617-622. (in Chinese)

YOON J H, KIM D Y, YOON K J. Visual Tracking Via Adaptive Tracker Selection with Multiple Features[M]. FITZGIBBON A, LAZEBNIK S, PERONA P, et al. Computer Vision-ECCV 2012, Berlin Heidelberg:Springer, 2012:28-41.

BHATTACHARYA S, BLUMENSATH T, MULGREW B, et al. Fast encoding of synthetic aperture radar raw data using compressed sensing[C]. Proceedings of the IEEE/SP 14th Workshop on Statistical Signal Processing, 2007. SSP'07, Madison, WI, USA:IEEE, 2007:448-452.

TAKHAR D, LASKA J N, WAKIN M B, et al. A new compressive imaging camera architecture using optical-domain compression[C]. Proceedings of the SPIE 6065, Computational Imaging IV, San Jose, CA:SPIE, 2006:606509.

LI H X, SHEN C H, SHI Q F. Real-time visual tracking using compressive sensing[C]. Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI:IEEE, 2011:1305-1312.

KALMAN R E. A new approach to linear filtering and prediction problems[J]. Transactions of the ASME-Journal of Basic Engineering, 1960, 82(Series D):35-45.

CHUI C K, CHEN G R. Kalman Filtering:with Real-Time Applications [M]. Berlin Heidelberg:Springer, 2009.

ACHLIOPTAS D. Database-friendly random projections:Johnson-Lindenstrauss with binary coins[J]. Journal of Computer and System Sciences, 2003, 66(4):671-687.

BARANIUK R, DAVENPORT M, DEVORE R, et al. A simple proof of the restricted isometry property for random matrices[J]. Constructive Approximation, 2008, 28(3):253-263.

ZHANG K H, ZHANG L, YANG M H. Real-Time Compressive Tracking[M]. FITZGIBBON A, LAZEBNIK S, PERONA P, et al. Computer Vision-ECCV 2012. Berlin Heidelberg:Springer, 2012:864-877.

HARRINGTON P. Machine Learning in Action[M]. Shelter Island, NY:Manning Publications, 2012.

DIACONIS P, FREEDMAN D. Asymptotics of graphical projection pursuit[J]. The Annals of Statistics, 1984:793-815.

HARE S, SAFFARI A, TORR P H S. Struck:structured output tracking with kernels[C]. Proceedings of 2011 IEEE International Conference on Computer Vision (ICCV), Barcelona:IEEE, 2011:263-270.

GRABNER H, LEISTNER C, BISCHOF H. Semi-Supervised on-line Boosting for Robust Tracking[M]. FORSYTH D, TORR P, ZISSERMAN A. Computer Vision-ECCV 2008. Berlin Heidelberg:Springer, 2008:234-247.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于卷积神经网络的响应自适应跟踪

基于Mamba-2的视频快照压缩成像重构方法

TFT-LCD表面Mura缺陷的AOI检测研究进展

基于主干增强和特征重排的反无人机目标跟踪

机器视觉在COB LED显示模组墨色采集、分类、排序中的应用