Image Denoising Using A Combined System Of Shearlets and Principle Component Analysis Algorithm  
  Authors : Linnet N Mugambi; Mwangi Waweru; Michael Kimwele

 

Although image denoising is widely studied, the effect of noise in image endures in image processing. Most of the existing research works have used the basic noise reduction through image blurring but without much impact. As researchers continue to focus on the subject, it is important to appreciate the need for effective denoising methods for quality images. While some methods have managed to denoise some types of noise, in the process they affect the image quality. This research intended to establish an approach for denoising images while maintaining the image quality. To create this approach, several denoising approaches and algorithms have been studied to determine their shortcomings and a combination of two, i.e. Shearlet Transform and PCA (Principle Component Analysis Algorithm was deemed viable in adding value to the existing denoising methods. The combination method increases the superiority of the observed image, subjectively and objectively.

 

Published In : IJCAT Journal Volume 3, Issue 1

Date of Publication : January 2016

Pages : 01 - 07

Figures :05

Tables : 03

Publication Link :Image Denoising Using A Combined System Of Shearlets and Principle Component Analysis Algorithm

 

 

 

Linnet N. Mugambi : is a career science teacher specializing in mathematics and physics with an experience spanning over 10 years. Linnet is an ongoing student of a Master of Science degree in computer systems at Jomo Kenyatta University of Agriculture and Technology (JKUAT). She holds a bachelor of education in science degree from Kenyatta University (2007) and a diploma in science education from Kenya science Teachers’ college (1998).

Waweru Mwangi : is an associate professor of information systems engineering at Jomo Kenyatta University of Agriculture and Technology (JKUAT) in the department of computing. He holds a bachelor’s degree from Kenyatta University, a master’s degree from Shanghai University (China) and a PhD degree from Hokkaido University (Japan). His main research interest includes Simulation and Modeling, Artificial Intelligence and Software Engineering.

Dr. Michael W. Kimwele : is a Lecturer in the Department of Computing, Jomo Kenyatta University of Agriculture and technology (JKUAT). He holds a BSc. Mathematics and Computer Science-First Class Honours from JKUAT (2002), a Masters in Information Technology Management from University of Sunderland-UK (2006) and a Doctorate in Information Technology from JKUAT (2012). At present, he is the Associate Chairman, Department of Information Technology, JKUATWestlands Campus.Dr. Kimwele has authored a commendable number of research papers in international/national conference/journals and also supervises postgraduate students in Computer Science/Information Technology. His research interests include Information systems management, Information Technology Security, Electronic Commerce, Mobile Computing, Social implications of computer applications, Human Computer Interaction, and Computer Ethics.

 

 

 

 

 

 

 

Denoising

Image

Blur Images

Psnr

The combined method gives better results both byhuman visual and by PSNR values Similarly, when the two methods are used the properties of shearlet to enhance and detect image edges and the property of PCA of preserving image structures improves the final mage.

 

 

 

 

 

 

 

 

 

[1] K. Guo, D. Labate, W. Lim, G. Weiss, and E. Wilson, “ Wavelets with composite dilations and their MRA properties”,Appl. Comput. Harmon. Anal., 20 , pp. 231-249 (2006). [2] D. Labate, W. Lim, G. Kutyniok and G. Weiss, and “ Sparse multidimensional representation using shearlets”, Wavelets XI (San Diego, CA, 2005), 254- 262, SPIE Proc. 5914 , SPIE, Bellingham, WA, 2005. [3] K. Guo and D. Labate, “ Optimally sparse multidimensional representation using shearlets,”, SIAM J Math. Anal., 39 pp.298-318, 2007. [4] K. Guo, D. Labate and W. Lim, “ Edge analysis and identification using the continuous shearlet transform”, to appear in Appl. Comput. Harmon. Anal. [5] G. Kutyniok and D. Labate, “ The construction of regular and irregular shearlet frames”, J. Wavelet Theory Appl., 1, pp.1-10, 2007. [6] R. R. Coifman, Y. Meyer, and M. V. Wickerhauser, “ Wavelet analysis and signal processing”’, Wavelets and their Applications, pp.153-178, 1992. [7] S. Mallat and S. Zhang, “ Matching pursuits with timefrequency dictionaries,” IEEE Trans. Signal Process., pp. 3397-3415, Dec. 1993. [8] A. Cohen, I. Daubechies and J. Feauveau, “ Biorthogonal bases of compactly supported wavelets,” Commun. on Pure and Appl. Math., 45:485-560, 1992. [9] A. Cunha, J. Zhou, and M. Do, “ Thenonsubsampledcontourlet transform: Theory, design, and applications,” IEEE Trans. Image Process., vol. 15, no. 10, pp. 3089-3101, Oct. 2006. [10] R. Coifman and D. Donoho, “ Translation invariant denoising,” Wavelets and Statistics, A. Antoniadis and G. Oppenheim,Eds. New York: SPringer-Verlag, pp. 125- 150, 1995. [11] D. Yang and X. Zhou, “Irregular wavelet frames on L2(Rn)”, Science in China Ser. A Mathematics, pp. 277-287, 2005. [12] P. Kittipoom, G. Kutyniok and W. Lim, “Construction of Compactly Supported Shearlets”, preprint. [13] G. Kutyniok and W. Lim, “Compactly Supported Shearlets are Optimally Sparse”, preprint.