Color Image Denoising Using Gaussian Multiscale Multivariate Image Analysis

Abstract:

Inspired by the human vision system, a new image representation and analysis model based on Gaussian multiscale multivariate image analysis (MIA) is proposed. The multiscale color texture representations for the original image are used to constitute the multivariate image, each channel of which represents a perceptual observation from different scales. Then the MIA decomposes this multivariate image into multiscale color texture perceptual features (the principal component score images). These score images could be interpreted as 1) the output of three color opponent channels: black versus white, red versus green and blue versus yellow, and 2) the edge information, and 3) higher-order Gaussian derivatives. Finally the color image denoising approach based on the models is presented. Experiments show that this denoising method against Gaussian filters significantly improves the denoising effect by preserving more edge information.

• Periodical: Applied Mechanics and Materials (Volumes 37-38)
• Pages: 248-252
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.37-38.248
• Online since: November 2010
• Authors: Dong Tai Liang
• Keywords: Color Image Denoising, Gaussian Multiscale Representations, Human Visual System, Multivariate Image Analysis

References

[1] A. Kumar and G. Pang: IEEE Trans. on Industry Applications, Vol. 38 (2002) No. 2, pp.425-440.

[2] J.J. Koenderink: Biological Cybernetics, Vol. 50 (1984) No. 5, pp.363-370.

[3] T. Lindeberg: Scale-Space Theory in Computer Vision (Kluwer/Springer, Boston, MA, 1994).

[4] T. Lindeberg: Journal of Applied Statistics, Vol. 21 (1994), No. 1-2, pp.225-270.

[5] B.M. ter Haar Romeny and L.M.J. Florack: IEEE International Workshop on Biologically Motivated Computer Vision, Seoul, Korea, (2000), pp.297-307.

DOI: 10.1007/3-540-45482-9_29

[6] T. Lindeberg: International Journal of Computer Vision, Vol. 30 (1998), No. 2, pp.1573-1405.