Face Image Denoising Method Based on Fourth-Order Partial Differential Equations

Abstract:

In order to improve the effect of face image denoising, this paper put forward several face image denoising methods based on partial differential equations, including P-M non-linear diffusion equations and fourth-order partial differential equations. We use those two methods by establishing non-linear diffusion equations and fourth-order anisotropic diffusion partial differential equation. The P-M non-linear diffusion denoising method can remove noise in intra-regions sufficiently but noise at edges can not be eliminated successfully and line-like structures can not be held very well.While the fourth-order partial differential equations denoising can retain the local detail characteristics of the original face image. Finally, through the experimental results we can see the effect of the fourth-order partial differential equations denoising is better, which makes the later face image processing more accurate and promotes the development of face image processing.

• Periodical: Advanced Engineering Forum (Volumes 6-7)
• Pages: 700-703
• DOI: https://doi.org/10.4028/www.scientific.net/AEF.6-7.700
• Online since: September 2012
• Authors: Weng Cang Zhao, Fan Wang
• Keywords: Face Image Denoising, Fourth-Order Partial Differential Equation (PDE), P-M Non-Linear Diffusion

References

[1] Chellappa M, Wilson C, Sirohey S, Human and machine recognition of faces: a survey, J. Proceeding of the IEEE. 83 (1995) 705-741.

DOI: 10.1109/5.381842

[2] Perona P, Malik J, Scale-space and edge detection using anisotropic diffusion, J. IEEE Trans on Pattern Anal Machine Intell. 12 (1990) 629-639.

DOI: 10.1109/34.56205

[3] Xinwu Liu, Lihong Huang, Zenyuan Guo, Adaptive fourth-order partial differential equation filter for image denoising, J. Applied Mathematics Letters. 24 (2011) 1282–1288.

DOI: 10.1016/j.aml.2011.01.028

[4] Linfang Wu, Naping Chen, Main models for image noise removal based on PDEs, J. Journal of Minjiang University. 27 (2006) 25-28.

[5] You Yu-Li, Kvech M, Four-order partial differential equations for noise removal, J. IEEE Trans on Image Processing. 9 (2000) 1723-1729.

[6] Kenneth R C, Digital image processing, Prentice Hall Internationl: Englewood Cliffs. (1988).

[7] Xiurui Geng, Suoping Li, Four-order partial differential equations used for image denoising,J. Gansu Industrial University Press. 28(2002) 256-259.