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...

光照突变环境下基于高斯混合模型和梯度信息的视频分割

DOI: 10.11834/jig.20071121

Keywords: 复杂背景分割,高斯混合模型,结构梯度互相关函数,光照突变

Full-Text   Cite this paper   Add to My Lib

Abstract:

基于高斯混合模型和帧间梯度信息提出了一种新的运动目标分割算法。首先,在利用亮度信息对背景建立自适应高斯混合模型的基础上,进行前景的粗分割;其次,由于视频信号的亮度和色彩分量随光照突变有较大的改变,导致大片背景的高斯模型产生错误匹配,误判为前景,为了提高高斯模型分割算法的鲁棒性,结合结构梯度互相关函数对分割结果进一步校正,能适应剧烈的光照变化;最后,利用数学形态学进行后处理,消除影子和孤立的噪声点。通过不同场景的运动分割实验结果表明,该算法在复杂背景和剧烈光照变化条件下具有较强的鲁棒性和较高的分割精度。

References

[1]  Lin Kai-yan,Wu Jun-hui,Xu Li-hong.Summary of segmentation methods in color image[J].Journal of Image and Graphics,2005,10(1):1~10.[林开颜,吴军辉,徐立鸿.彩色图像分割方法综述[J].中国图象图形学报,2005,10(1):1~10.]
[2]  Wei Hong-bo,Lü Zhen-su,Jiang Tian-zi,et al.Survey of image segmentation techniques[J].Gansu Science Ransaction,2004,16(2):19~24.[魏弘博,吕振肃,蒋田仔等.图像分割技术纵览[J].甘肃科学学报,2004,16(2):19-24.]
[3]  Arandjelovi,Cipolla R.Incremental learning of temporally-coherent Gaussian mixture models[A].In:Proceedings of British Machine Vision Conference[C],Oxford,UK,2005:759~768.
[4]  Stauffer C,Grimson W.Learning patterns of activity using real-time tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):747~757.
[5]  Paul Rosin.Thresholding for change detection[J].Computer Vision and Image Understanding,2002,88(2):79~95.
[6]  Wang Ze-bing,Yang Chao-hui.Research of segmentation techniques in color image[J].Journal of Digital Television and Digital Video,2005,4:20~24.[王泽兵,杨朝晖.彩色图像分割技术研究[J].数字电视与数字视频,2005,4:20~24.]
[7]  Li Li-yuan,Huang Wei-min,Gu Yu-hua,et al.Statistical Modeling of Complex Backgrounds for Foreground Object Detection[J].IEEE Transactions on Image Processing,2004,13(11):1459~1472.
[8]  Michael Harville,Gaile Gordon,John Woodfill.Foreground segmentation using adaptive mixture models in color and depth[A].In:IEEE Proceedings of IEEE Workshop on Detection and Recognition of Events in Video[C],Vancouver,Canada,2001:3~11.
[9]  Stauffer C,Grimson W.Adaptive background mixture models for real-time tracking[A].In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition[C],Fort Collins,Colorado,1999:246~252.
[10]  Chen Bai-sheng,Lei Yun-qi,Li Wang-wei.A novel background model for real-time vehicle detection[A].In:ICSP\'04 Proceedings of IEEE 2004[C],Beijing,China,2004,(4):1276~1279.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133