Optimal Denoising of an Image using Anscombe Transformation Based Image Stabilization
Sophia Comaneci.J1, K.John Peter2
1Sophia Comaneci .J, Computer Science and Engineering, Vins Christian College of Engineering, Kanyakumari, India.
2Mr. K. John Peter, Computer Science and Engineering, Vins Christian College of Engineering, Kanyakumari, India.
Manuscript received on May 17, 2012. | Revised Manuscript received on June 22, 2012. | Manuscript published on June 30, 2012. | PP: 246-250 | Volume-1 Issue-5, June 2012. | Retrieval Number: E0497061512/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper proposes an effective inversing of the anscombe transformation with the help of adaptive bilateral image denoising algorithm. The Poisson noise removal is carried out into three steps. They are First, Image pre-processin,. Second, image denoising and Third, Image retrieval. In image pre-processing the images of any format can be got as input they are then converted into gray scale images for ease of functions and this paper uses anscombe transform to stabilize the image to a constant intensity level. This is very helpful in determining the noise at low counts. For image denoising, Multiscale variance stabilizing transform is the technique that is proposed to denoise the image. Now the noisy pixels in the images are removed. This paper also proposes a similar neighborhood function that is essential for filling the noisy pixels with the help of non-local means of similar neighbors. This is suitable for overall adjustment of the image. But in the case of texture images this technique is not applicable and in that condition the technique proposed is bilateral transformation of texture images. For this we use Bilateral image denoising and PCA analysis. This paper also proposes an approach to determine the best among the two processes in terms of performance and efficiency. Next step is very crucial because the application of inverse transformation is an critical factor. The inverse transform that is proposed in this paper is minimum mean square error method. This results in retrieval of an image with efficient filtering and inversing functions.
Keywords: Anscombe transform, MS-VST, Bilateral denoising, PCA analysis, MMSE