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...

一种基于区域收缩的运动分割算法

DOI: 10.11834/jig.20080819

Keywords: 运动分割,最大后验边缘概率,区域收缩

Full-Text   Cite this paper   Add to My Lib

Abstract:

运动分割需要估计出每个运动的运动模型参数和运动支持区。为准确地确定运动支持区,在最大后验边缘概率(MPMMAP)算法的基础上,提出了一种新的基于区域收缩的运动分割算法,用于确定运动支持区。该算法先以属于某种运动概率最大的像素为备选像素,然后通过区域收缩选定备选像素密度最大的区域为支持区范围。此外,还提出了一种新的运动模型参数初值的估计方法,并将差分图像包围盒的确定和区域收缩相结合,用于运动模型参数的初值估计,该方法先估计运动区域,再估计运动模型参数,并通过运动分解、合并和消亡来获得准确的运动个数。实验结果表明,该方法是有效的。

References

[1]  Weiss Y. Smoothness in layers: motion segmentation using nonparametric mixture estimation [ A ]. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], San Juan, Puerto Rico, 1997 : 520 - 527.
[2]  Weiss Y, Adelson E H. A unified mixture framework for motion segmentation : incorporing spatial coherence and estimating the number of model[A]. In: Proceedings of IEEE Confercence on Computer Vision and Pattern Recognition[C] ,Vienna, Austria, 1996:321-326.
[3]  Vasconcelos N, Lippman A. Empirical Bayesian motion segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(2):217-221.
[4]  Zhang J. The mean field theory in EM procedures for Markov random fields[J]. IEEE Transactions on Signal Process, 1992, 40 (10) : 2570 - 2583.
[5]  Celeux G, Forbes F. EM procedures using mean field-like approximations for Markov model-based image segmentation [J]. Pattern Recognition, 2003, 36( 1 ) : 131 - 144.
[6]  Marroquin J L, Velasco F, Rivera M, Nakamura M. Gauss-markov measure field model for low-level vision [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001,23 (4) : 337 - 348.
[7]  Wang J Y A, Adelson E H. Representing moving images with layers[J]. IEEE Transactions on Image Processing, 1994, 3(5) : 625 -638.
[8]  Elias D P, Kingsbury N G. The recovery of a near optimal layer representation for an entire image sequences[ A]. In: Proceedings of IEEE Confercence on Image Processing[C] ,Washington, DC, USA, 1997 : 735 -738.
[9]  Zhang W, Fang X. Moving vehicles segmentation based on Bayesian framework for Gaussian motion model [J]. Pattern Recognition Letters, 2006, 27(9): 956-967.
[10]  Zhang J. The mean field theory in EM procedures for Markov random fields[J]. IEEE Transactions on Image Processing, 1993, 40 (10) : 2570 - 2583.
[11]  Jaakkola T S, Jordan M I. Improving the mean field approximation via the use of mixture distributions [ A ]. In: Jordan M I. ( Ed. ), Learning in Graphical Models [C], Dordrecht: MA, USA: Kluwer Academic Publishers, 1998:163 - 173.
[12]  Calderon F, Marroquin J L. The MPM-MAP algorithm for motion segmentation [J]. Computer Vision and Image Understanding, 2004, 95(2) : 165 -183
[13]  Black M, Anandan P. The robust estimation of multiple motions: parametric and piecewise-smooth flow fields [J]. Computer Vision and Image Understanding, 1996,63 ( 1 ) :75 - 104.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133