Zeng Rui
School of Electro-mechanical and Information Technology, Yi Wu Industrial and Commercial College, Yi Wu, China
Wang Ying-yan
School of Electro-mechanical and Information Technology, Yi Wu Industrial and Commercial College, Yi Wu, China
ABSTRACT
We design is based on the shape of the outline track ICA energy model and derive the evolution curve gradient flow equations. The first part of the energy functional is based on a priori information about the shape of the energy term, used to constrain the active outline. Active Outline and expectations in the evolution of the shape of constantly comparing it with apriori shape model to match. The second part is the image energy term, the candidate target area by minimizing the covariance with the template from the background region candidate while maximizing the covariance with the template from the evolution of the evolution of the curve to the target location. The third part is the length of the term, limiting the evolution of the curve and make a smooth curve. Validated based on ICA shape outline track algorithm is effective, we have a different image sequence of human walking on the test. Experimental results show that our proposed algorithm processing including noise, illumination variation and complexity of issues such as the background image is robust.
PDF References Citation
How to cite this article
Zeng Rui and Wang Ying-yan, 2013. Outline Track Algorithm Based on ICA Shape Feature. Information Technology Journal, 12: 4203-4211.
DOI: 10.3923/itj.2013.4203.4211
URL: https://scialert.net/abstract/?doi=itj.2013.4203.4211
DOI: 10.3923/itj.2013.4203.4211
URL: https://scialert.net/abstract/?doi=itj.2013.4203.4211
REFERENCES
- Correa, N., T. Adali, Y.O. Li and V.D. Calhoun, 2005. Comparison of blind source separation algorithms for FMRI using a new Matlab toolbox: GIFT. IEEE International Conference on Acoustics, Speech and Signal Processing, Volume 5, March 18-23, 2005, Philadelphia, PA., USA., pp: v-401-v-404.
CrossRef - Hosseini, S. and Y. Deville, 2004. Blind maximum likelihood separation of a linear-quadratic mixture. Proceedings of the 5th International Conference on Independent Component Analysis and Blind Signal Separation, September 22-24, 2004, Granada, Spain, pp: 694-701.
CrossRef - Hu, D., L. Yan, Y. Liu, Z. Zhou, K.J. Friston, C. Tan and D. Wu, 2005. Unified SPM-ICA for fMRI analysis. Neuroimage, 25: 746-755.
CrossRef - Ku, W., R.H. Storer and C. Georgakis, 1995. Disturbance detection and isolation by dynamic principal component analysis. Chemometr. Intell. Lab. Syst., 30: 179-196.
CrossRef - Lee, J.M., C. Yoo and I.B. Lee, 2004. Statistical monitoring of dynamic processes based on dynamic independent component analysis. Chem. Eng. Sci., 59: 2995-3006.
CrossRef - Lu, C.J., C.M. Wu, C.J. Keng and C.C. Chiu, 2008. Integrated application of SPC/EPC/ICA and neural networks. Int. J. Prod. Res., 46: 873-893.
CrossRef - Nomikos, P. and J.F. MacGregor, 1995. Multivariate SPC charts for monitoring batch processes. Technometrics, 37: 41-59.
Direct Link - Shao, R., F. Jia, E.B. Martin and A.J. Morris, 1999. Wavelets and non-linear principal components analysis for process monitoring. Control Eng. Pract., 7: 865-879.
CrossRef - Sun, R. and F. Tsung, 2003. A kernel-distance-based multivariate control chart using support vector methods. Int. J. Prod. Res., 41: 2975-2989.
CrossRefDirect Link