一种改进的一致性扩散图像增强方法

喻罡; 张红梅; 卞正中; 郭佑民

一种改进的一致性扩散图像增强方法

喻罡^{①;张红梅},
张红梅,
卞正中,
郭佑民

计量
- 文章访问数: 2320
- HTML全文浏览量: 84
- PDF下载量: 832
- 被引次数: 0
出版历程
- 收稿日期: 2004-04-26
- 修回日期: 2005-01-24
- 刊出日期: 2005-09-19

An Improved Coherence Diffusion Method for Image Enhancement

摘要

摘要: 一致性扩散是分析图像中定向结构的重要的预处理步骤,现有的一致性扩散图像增强方法难以识别图像中的弱边界.本文提出了一种改进的扩散方法,该方法构建了一种结合二阶方向导数信息的结构张量,能够精确分析图像中的复杂弱边界,同时与经典结构张量互为补充以检测图像的强边界.在此基础上设计了新的扩散张量,并采用了基于(加性分裂算子AOS)策略的数值计算方法.通过对比实验证明,该文算法的效果良好并具有很高的效率,实现了在去除噪声的同时精确保护图像中的强弱边界.
- 一致性图像增强; 结构张量; 扩散张量; 二阶方向导数; AOS策略
Abstract: Coherence diffusion is an important preprocessing step for analyzing oriented structures in the image. Previous coherence diffusion methods for image enhancement could not recognize weak boundaries. In this paper, an efficient diffusion approach is presented. A new structure tensor integrating the second-order directional derivative information is designed, which can precisely analyze complex weak edges in the image. By combining the proposed structure tensor and the classical one as complementary descriptor, the improved diffusion tensor is constructed to detect strong edges simultaneously. Furthermore, parallel AOS （Additive Operator Splitting） scheme is applied to implement numerical solution, which is faster than usual numerical approach. Promising experimental results of several real images demonstrate that the new diffusion approach can preserve important strong edges and weak edges precisely while removing the noise.

HTML全文

参考文献(1)

Weickert J. A review of nonlinear diffusion filtering. Proceedings of the First International Conference on Scale-Space Theory in Computer Vision, London UK, 1997, 1252:3 - 28.[2]Perona P, Malik J. Scale-space and edge detection using anisotropic diffusion[J].IEEE Trans. on Machine Intelligence,Berkey USA.1990, 12 (7):629-[3]Saito T, Harada H, Satsumabayashi J, Kumatsu T. Color image sharpening based on nonlinear reaction-diffusion. 2003International Conference on Image Processing, Yokohama,Japan, Sept. 2003, 3:389 - 392.[4]Rudin L I, Osher S, Fatemi E. Nonlinear total variation based noise removal algorithms[J].Physica D.1992, 60(4):259-[5]Weisenseel R A, Karl W C, Castanon D A. A region-based alternative for edge-preserving smoothing. Proceedings 2000International Conference on Image Processing, Vancouver,Canada, Sept. 2000, 3:778 - 781.[6]Diewald U, Preusser T, Rumpf M, Strzodka R. Diffusion models and their accelerated solution in image and surface processing.Proceeding of Algoritmy 2000, Acta Math Univ Comenianae,2001, LXX: 15 - 31.[7]Chen Y, Levine S E. Image recovery via diffusion tensor and time-delay regularization[J].Journal of Hsual Communication and Image Representation.2002, 13(3):156-[8]Weickert J. Coherence-enhancing diffusion of color images[J].Image and Vision Computing.1999, 17(3):201-[9]Bakalexis S A, Boutalis Y S, Mertzios B G. Edge detection and image segmentation based on nonlinear anisotropic diffusion.2002 14th Intemational Conference on Digital Signal Processing,Xanthi, Greece, July 2002, 2:1203 - 1206.[10]Cottet G H, Germain L. Image processing through reaction combined with nonlinear diffusion[J].Math Comp.1993, 61(10):659-[11]Brox T, Weickert J. Nonlinear matrix diffusion for optic flow estimation. Pattern Recognition, 24th DAGM-Symposium, 2002:446 - 453.[12]Weickert J, Zuiderveld K J. Parallel implementations of AOS schemes: A fast way of nonlinear diffusion filtering. IEEE International Conference on Image Processing, Santa Barbara,Oct. 1997, 3:396 - 399.

施引文献

资源附件(0)

访问统计