全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于Cauchy分布的红外视频运动目标检测

Keywords: 红外,智能视觉监视,变化检测,背景建模,柯西分布,梯度

Full-Text   Cite this paper   Add to My Lib

Abstract:

基于柯西分布统计模型和梯度信息,提出了一种用于热成像(红外)图像序列中行人目标检测的统计背景剔除方法.它不依赖传统红外行人检测方法的"热场"(hot-spot)假设(即认为人体区域总比周围环境热),而是用一种快速动态背景的重建方法建立初始化背景Gaussian模型,简化了Cauchy分布的参数估计问题.基于Gauchy分布的背景剔除被首先用来识别局部兴趣区域(Regions-Of-Interest,ROI).在每个兴趣区域将当前图像和背景图像的梯度信息组合成一个梯度图.对梯度图进行各向异性滤波和运用最大类间方差法进行自适应阈值处理后获得行人目标区域.经多组实验及与其他方法比较,结果表明提出的方法对热成像红外视频人体目标的位置和形状实现了较精确的检测,对环境、时间变化、极性翻转、"晕轮"效应等产生的影响,以及背景中杂波的干扰具有较好的鲁棒性.

References

[1]  LING Jian-Guo,LIU Er-Qi,LIANG Hai-Yan,et al.Infra-red target extraction method based On kernel density estima-tion[J].J.Infrared Millim.Waves(凌建国,刘尔琦,梁海燕,等.基于核密度估计的红外目标提取方法.红外与毫米波学报),2006,25(6):434-438.
[2]  Bhanu B,Han J.Kinematic-based human motion analysis in infrared sequences[C].Proceedings of Sixth IEEE Workshop On Applications of Computer Vision,2002:208-212.
[3]  Nanda H.,Davis L.Probabilistic template based podestrian detection in infrared videos[C].IEEE Proc.Intel.Vehicles Syrup.2002,1:15-20.
[4]  Xu Feng-Liang,Hu Xia,Fujimura K.Pedestrian detection and tracking with night vision[J].IEEE Transactions On Intelligent Transportation Systems,2005,6(1):63-71.
[5]  Davis J,Sharma V.Robust background-subtraction for per-son detection in thermal imagery[C].IEEE internatonnal workshop on object tracking and classification beyond the visi-ble spectrum,Washington D.C,July,2004,8:128-135.
[6]  Til Aach.Statistical model-based change detection in mov-ing video[J].signal Processing.1993,31:165-180.
[7]  Skifstad K.Jain R.Illunination independent chanse detec-tion for real world image sequence[C].CVIP,1989,46(3):387-399.
[8]  REN Ying,CHUA Ching-seng.Motion detection from time-varied background[J].International Journal of Image and Graphics,2002,2(2):163-178.
[9]  Tsihrintzis G A,Nikias C L.Fast estimation of the parame-ters of alpha-stable impulsive interference using asymptotic ex-treme value theory[C].ICASSP-95,9-12 May 1995,3:1840-1843.
[10]  MING Ying,JIANG Jing-Jue,YANG Jian.Statistical change detection based on Cauchy distribution for visual surveillanee[J].IET International Conference on Visual Information Engineering.2006:559-564.
[11]  U Yu-Qi,HE Pin.The theory of probability and statistics[M].Beijing:National defense industry publishing com-pany.(李裕奇,何平编审.概率论与数理统计.北京:国防工业出版社),2001:130-131.
[12]  Black M J,Sapiro G,Marimont D H,et al.Robust aniso-tropic diffusion[J].IEEE transactions on image process-ing,1998,7(3):421-432.
[13]  Otsu N.A threshold selection method from gray-level his-togramg[J].IEEE Transactions on Systems,Man,and Cybernetics.1979,9(1):62-66.
[14]  Conaire C o,Cooke E,O\'Counor N,et al.Background modelling in infrared and visible spectrum video for people tracking[C].Proceedings of the 2005 IEEE computer soci-ety conference On computer vision and pattern recognition (CVPR\'05),2005.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133