Image De-Noising Based on Improved Data-Adaptive Kernel Regression Method

Zhen Ping Qiang; Xu Chen; Hong Lin; Tong Lin Zhao

doi:10.4028/www.scientific.net/AMR.532-533.1359

Paper Titles

Research for the SSD-Based Storage Technology of Mass Remote Sensing Image
p.1339

Knowledge Integration for Analyzing ChIP-seq
p.1344

Development of Vibration Data Processing Program
p.1349

Harmonic Monitoring Based on BP Neural Network
p.1354

Image De-Noising Based on Improved Data-Adaptive Kernel Regression Method
p.1359

Research of Query Optimization Based on Improved Quantum Particle Swarm Optimization Algorithm in Distributed Database
p.1365

Adaptive K-Means Algorithm with Dynamically Changing Cluster Centers and K-Value
p.1373

Blind Source Separation Based on Non-Gaussianity
p.1378

Decision Tree Based Rules Redistribution Algorithm for IP Monitoring of Lawful Interception on IPv4 Networks
p.1384

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 532-533Image De-Noising Based on Improved Data-Adaptive...

Image De-Noising Based on Improved Data-Adaptive Kernel Regression Method

Abstract:

This paper proposes a novel method for image de-noising, the algorithm is improved the data-adaptive kernel regression method. The process of each pixel is: first determine whether the pixel is on boundary, for the pixels on the edge to establish the kernel which shape is adaptive with the boundary, and then use iterative process for de-noising. For non-boundary pixels, use the data-adaptive iterative kernel regression method. Experiments have shown promising results in image de-noising; the algorithm is able to filter out the high-frequency noise of image while it retains the details of the image characteristics.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 532-533)

Pages:

1359-1364

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.532-533.1359

Citation:

Cite this paper

Online since:

June 2012

Authors:

Zhen Ping Qiang, Xu Chen, Hong Lin, Tong Lin Zhao

Keywords:

Data-Adaptive, De-Noising, Edge Adaptive, Kernel Regression

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] M. Lindenbaum, M. Fischer, and A. Bruckstein. On gabor contribution to image enhancement, Pattern Recognition, 1994, 27: 1–8.

DOI: 10.1016/0031-3203(94)90013-2

Google Scholar

[2] P. Perona and J. Malik. Scale space and edge detection using anisotropic diffusion, IEEE Trans. Patt. Anal. Mach. Intell., 1990, 12: 629–639.

DOI: 10.1109/34.56205

Google Scholar

[3] L. Rudin, S. Osher, and E. Fatemi. Nonlinear total variation based noise removal algorithms, Physica D, 1992, 60: 259–268.

DOI: 10.1016/0167-2789(92)90242-f

Google Scholar

[4] L. Yaroslavsky. Digital Picture Processing - An Introduction. Springer Verlag, (1985).

Google Scholar

[5] M. P. Wand and M. C. Jones, Kernel Smoothing, ser. Monographs on Statistics and Applied Probability. New York: Chapman & Hall, (1995).

Google Scholar

[6] Hiroyuki Takeda, Sina Farsiu and Peyman Milanfar. Kernel Regression for Image Processing and Reconstruction. IEEE transactions on image processing, 2007, VOL. 16, NO. 2, FEBRUARY.

DOI: 10.1109/tip.2006.888330

Google Scholar

[7] Weinberger.K. and Tesauro.G. Metric Learning for Kernel Regression. Eleventh International Conference on Artificial Intelligence and Statistics, Omnipress, Puerto Rico, 2007, pp.608-615.

Google Scholar

[8] Feng X andM ilan far P. Multiscale principal components analysis for image local orientation estimation, Presented at the 36th As ilom arCon. f Signals, System s and Com puters, Pacific Grove, CA, Nov. 2002. Fig. 4, Block diagram of the method Fig. 5 Improved Data-Adaptive Kernel Regression Method. (a) Original image. (b) Noisy image. (c) Smooth processed of image (b). (d) Gradient caculated of image (b). (e) Edge detection of image (c). (f) Kernel regression based on image (b), (d) and (e). Fig. 6 Comparison of denoising algorithm. (a) Original image. (b) Noisy image. (c) Mean filtering method. (d) median filtering method. (e) Data-adaptive kernel regression. (f) The proposed improved data-adaptive kernel regression. Fig. 2, Control the shape of the kernel.

DOI: 10.4028/www.scientific.net/amr.532-533.1359

Google Scholar