非凸泛函的图像分解

doi:10.3969/j.issn.1001-2400.2013.02.012

Abstract

Abstract:

This paper proposes a new model for image decomposition by non-convex functional minimization. Instead of using the Banach norm as the fidelity term, we use the integral of the square of residual component divided by its gradient as the fidelity term. This non-convex fidelity term has a very low value for the texture image and a high value for the geometric image, so it is appropriate for image decomposition. The gradient descent procedure is used to solve the proposed minimization problem, which leads to evolving a new nonlinear second-order partial differential equation(PDE) to a steady state. Compared with the total variation minimization(TV) model and the fourth-order PDE(OSV) model, the proposed nonlinear second-order PDE maintains many more sharp edges, so the texture part has less cartoon information. Experimental results also demonstrate that our model performs better than the standard TV and OSV models in image decomposition.

Key words: image decomposition, total variation minimization, functional minimization, non-convex functional

CLC Number:

TP751

BAI Jian;FENG Xiangchu. Image decomposition using the non-convex functional[J].J4, 2013, 40(2): 67-71+137.

References

［1］ Rudin L, Osher S, Fatemi E. Nonlinear Total Variation Based Noise Removal Algorithms ［J］. Physica D, 1992, 60(1-4): 259-268.
［2］ Gillers J, Meyer Y. Properties of BV-G Structures+Textures Decomposition Models Application to Road Detection in Satellite Images ［J］. IEEE Trans on Image Processing, 2010, 19(11): 2793-2800.
［3］ Meyer Y. Oscillating Patterns in Image Processing and Nonlinear Evolution Equations ［R］. Boston: American Mathematical Society, 2001.
［4］ Vese L, Osher S. Modeling Textures with Total Variation Minimization and Oscillating Patterns in Image Processing ［J］. Journal of Scientific Computing, 2003, 19(1/3): 553-572.
［5］ Osher S, SoléA, Vese A. Image Decomposition and Restoration Using Total Variation Minimization and the H^-1 Norm ［J］. Multiscale Model and Simulation, 2003, 1(3): 349-370.
［6］ Bai J, Feng Xiangchu. Image Denoising and Decomposition Using Non-Convex Functional ［J］. Chinese Journal of electronics, 2012, 12(1): 102-106.
［7］ Bai J, Feng Xiangchu. A Novel Nonlinear Parabolic-Hyperbolic Equation for Image Decomposition ［C］//2011 IEEE 3rd International Conference on Signal Processing Systems. Beijing: IEEE, 2011: 41-45.
［8］ Aujol J, Chambolle A. Dual Norms and Image Decomposition Models ［J］. International Journal of Computer Vision, 2005, 63(1): 85-104,.
［9］ Budeas A, Le T, Morel J, et al. Fast Cartoon+Texture Image Filters ［J］. IEEE Trans on Image Processing, 2010, 19(8): 1978-1986.
［10］ Scherzer O, Grasmair M, Grossauer H, et al. Variational Methods in Imaging ［M］. New York: Springer-Verlag, 2009.
［11］王旭东, 冯象初, 白键. 基于结构张量场拟合的图像恢复方法［J］. 西安电子科技大学学报, 2011, 38(6): 68-74.
Wang Xudong, Feng Xiangchu, Bai J. Novel Image Restoration Method with Structure Tensor Fitting ［J］. Journal of Xidian University, 2011, 38(6): 68-74.
［12］姚晓莉, 冯象初, 李亚峰. 去除乘性噪声的主成分分析算法［J］. 光子学报, 2011, 40(7): 1031-1035.
Yao Xiaoli, Feng Xiangchu, Li Yafeng. Principal Component Analysis Method for Muitiplicative Noise Removal ［J］. Acta Photonica Sinica, 2011, 40(7): 1031-1035.

[1]	KUANG Tao;HUANG Liyu;ZHONG Yufang;LI Chao. Improved compressive sensing algorithm for CT image reconstruction with incomplete projection data [J]. J4, 2015, 42(4): 95-99+113.
[2]	BAI Jian;FENG Xiangchu. Model based on the integro-differential equation for multiplicative noise removal [J]. J4, 2013, 40(3): 132-138.
[3]	. Image Decomposition Using Non-convex Functional [J]. J4, 2013, 40(2): 67-71.
[4]	LU Cheng-wu1;2. Multiscale decomposition of image under (B_V,E) frame [J]. J4, 2009, 36(1): 171-176.

Image decomposition using the non-convex functional

PDF (PC)

Like

Knowledge

Cited

Abstract

Cite this article

share this article

References

Related Articles 4

Metrics

Comments

Recommended 10