基于相位一致性的实时压缩跟踪方法

针对基于压缩感知的目标跟踪算法在跟踪过程中,光照剧烈变化引起跟踪不稳定或跟踪失败的问题, 本文提出了一种基于相位一致性的改进跟踪方法.该方法利用相位一致性图像特征对光照变化不敏感的特点, 首先对样本搜索区域内的图像进行相位一致性变换, 然后再提取变换后相位一致性图像的特征,将其用于分类器中来确定目标位置.实验结果表明, 该方法在目标受到光照剧烈变化影响的情况下具有很强的适应性, 在目标大小为50 pixel×55 pixel时平均处理帧频可达22 fps.与已有基于压缩感知跟踪算法相比, 该算法在光照变化剧烈的情况下仍具有很好的鲁棒性,而且在目标尺度和纹理发生一定变化的情况下跟踪稳定.

Abstract

The target tracking algorithm based on compressive sensing can cause the instability or failure in the tracking process when the illumination changes drastically.To deal with such problem , a developed tracking algorithm based on phase congruency was proposed,which is insensitive to the illumination.The phase congruency transformation of the image in the search area is calculated firstly, then the features extracted from the transformed image are used in the classifier to determine the target′s location.Experimental results show that the proposed method has a strong adaptability when the target has a drastic variation in the illumination, and the average frame rate can reach 22 fps when the target size is 50 pixel×55 pixel.Compared with the tracking algorithm based on compressive sensing, the proposed algorithm still has a very good robustness to the drastic variation of the target′s illumination.Besides, to a certain extent, the tracking is also stable to scale and textures change.

参考文献

[1] 孟勃,韩广良,毛雯蓓.采用栅格特征及可视化技术感知跟踪中目标的遮挡状态［J］.光子学报,2013, 42(1):98-103.

MENG Bo, HAN Guang-liang, MAO Wen-bei.Grid reature and visualization for occlusion perception in object tracking［J］.Acta Photonica Sinica,2013, 42(1):98-103.

[2] HE Sheng-feng,YANG Qing-xiong, RYNSON W H,et al.Visual tracking via locality sensitive histogram［C］.Computer Vision and Pattern Recognition, 2013: 2427-2434.

[3] WU Yi, LIM Jong-woo, YANG Ming-hsuan.Online object tracking : a benchmark ［C］.Computer Vision and Pattern Recognition, 2013: 2411-2418.

[4] DONOHO D L.Compressed sensing ［J］.Information Theory, 2006, 52(4): 1289 -1306.

[5] ZHANG K H, ZHANG L,YANG M H.Real-time compressive tracking ［C］.European Conference on Computer Vision, 2012:866 -879.

[6] ZHANG Kai-hua,SONG Hui-hui.Real-time visual tracking via onlineweighted multiple instancelearning［J］.Pattern Recognition, 2013,46:397–411.

[7] 朱秋平,颜佳,张虎,等.基于压缩感知的多特征实时跟踪［J］.光学精密工程,2013,21(2):437-444.

ZHU Qiu-ping, YAN Jia, ZHANG Hu, et al.Real-time trackingusing multiple features based on compressive sensing［J］.Opticsand Precision Engineering, 2013, 21(2):437-444.

[8] 陈涛,李正炜,王建立,等.应用压缩传感理论的单像素相机成像系统［J］.光学精密工程,2012,20(11):2523-2530.

CHEN Tao, LI Zhen-wei, WANG Jian-li, et al.Imaging system of single pixel camera based on compressed sensing［J］.Opticsand Precision Engineering, 2012,20(11):2523-2530.

[9] 朱明,高文,郭立强.压缩感知理论在图像处理领域的应用［J］.中国光学,2011,4(5):441-447.

ZHU Ming, GAO Wen, GUO Li-qiang.Application of compressed sensing theory in image processing［J］. Chinese Optics, 2011, 4(5):441-447.

[10] 符冉迪, 金炜, 叶明, 等.抗混叠轮廓波域采用压缩感知的云图融合方法［J］.光子学报, 2011, 40(6): 955-960.

FU Ran-di, JIN Wei , YE Ming,et al.Cloud image fusion using compressed sensing in aliasing – free contourlet domain［J］. Acta Phtonica Sinica,2011, 40(6):955-960.

[11] ACHLIPTAS D.Datebase-friendly random projection :Johnson-Lindenstrauss with binary coins［J］.Journal of Computer and System Sciences, 2003, 66(6):671-687.

[12] DIACONIS P,FREEDMAN D.Asymptotics of graphical projection pursuit［J］. The Annals of Statistics, 1984,12(3): 793-815.

[13] 罗海波,史泽林,陈永红.相位一致性图像及其在目标跟踪中的应用［J］.光子学报, 2010, 39(3): 547-552.

LUO Hai-bo, SHI Ze-lin,CHEN Yong-hong.Phase congruency image and its application in target tracking［J］.Acta Phtonica Sinica, 2010,39(3):547-552.

[14] 苑玮琦,范永刚,柯丽.相位一致性和对数Gabor滤波器相结合的掌纹识别方法［J］.光学学报,2010,30(1):147-152.

YUAN Wei-qi,FAN Yong-gang,KE Li.Palmprints recognitionmethod based on the phase consistency combined Log-Gabor filter ［J］.Acta Optica Sinica, 2010, 30(1):147-152.

[15] 郭龙源,卢阿丽,杨静宇.基于相位一致性变换的红外图像区域匹配算法研究［J］.红外与毫米波学报, 2009, 28(1):35-45.

GUO Long-yuan, LU A-Li,YANG Jing-yu.Infrared image region matching algorithms based on phase congruency transformation ［J］.Journal of Infrared and Millimeter Waves, 2009,28(1):35-45

[16] KOVESI P.Image feature from phase congruency ［J］.Videm:Journal of Computer Vision Research, 1999, 1(3):1-27.

[17] XU Jia , LU Hu-chuan, YANG Ming-hsuan.Visual tracking via adaptive structural local sparse appearance mode ［C］.Computer Vision and Pattern Recognition, IEEE Computer Society Conference, 2012:1822-1829．

[18] ABENKO B,YANG Ming-hsuan,BELONGIE S.Robust object tracking with online multiple instance learning ［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011,33:1619-1632.

[19] ADAM A,RIVLIN E,SHIMSHONI I.Robust fragments-based tracking using the integral histogram［C］.Computer Vision and Pattern Recognition,2006:798-805.

[20] COLLINS R,YANX L,LEORDEANU M.Online selection of discriminative tracking features ［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1631-1643.

张雷, 王延杰, 何舒文. 基于相位一致性的实时压缩跟踪方法[J]. 光子学报, 2014, 43(8): 0810003. ZHANG Lei, WANG Yan-jie, HE Shu-wen. Real-time Compressive Tracking Method Based on Phase Congruency[J]. ACTA PHOTONICA SINICA, 2014, 43(8): 0810003.

基于相位一致性的实时压缩跟踪方法

关于本站 Cookie 的使用提示

全站搜索

基于相位一致性的实时压缩跟踪方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索