Image Denoising and Enhancement using Multilevel 2-D Dwt Lifting
Mahesh N. Javalkar1, Gouri G. Uppin2
1Mahesh N. Javalkar, Lecturer, Department of Electronics and Communication Engineering, Maratha Mandal Polytechnic, Belgaum, (Karnataka) India.
2Gouri G. Uppin, Lecturer, Department of Electronics and Communication Engineering, Maratha Mandal Polytechnic, Belgaum, (Karnataka) India.
Manuscript received on 20 January 2016 | Revised Manuscript received on 30 January 2016 | Manuscript published on 30 January 2016 | PP: 9-14 | Volume-4 Issue-6, January 2016 | Retrieval Number: F1514014616©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© 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: In general, images are often corrupted by noise in the procedures of image acquisition and transmission. The noise may seriously affect the performance of image processing techniques . Hence image denoising and enhancement plays an important role in the field of image processing. The Wavelet Transform provides a scale based decomposition . For discrete time signals , Discrete Wavelet Transform (DWT) is implemented by two methods , the convolution method and the lifting scheme method . The basic idea is done by filtering the input signal with a low pass filter and a high pass filter and downsampling the outputs by a factor 2 . The lifting scheme is an efficient method of wavelet transform and is far better than the convolution method because of its advantages like faster implementation of wavelet transform , requires lesser number of computations , allows fully in-place calculation and reversible integer wavelet transform. The lifting scheme can be applied forwardly to enhance or denoise the image and it can further be applied inversely to get back the original image. In this paper the 2-dimensinal lifting based discrete wavelet transform (2-D DWT) method is implemented for image denoising and enhancement . The 2-D DWT lifting scheme algorithm has been implemented using MATLAB program for both modules , Forward Discrete Wavelet Transform (FDWT) and Inverse Discrete Wavelet Transform (IDWT) to determine the Mean Square Error (MSE) and the Peak Signal to Noise Ratio (PSNR) for the retrieved image . To implement denoising different noisy images are taken and denoised using 2-D DWT lifting scheme. The results are compared in terms of MSE, PSNR and execution time values . To verify enhancement the proposed lifting based DWT method is compared with Histogram Equilization(HE) method. The results show much more improved contrast enhancement by lifting based DWT as compare to the HE method . The parameter comparisons like MSE (reduced to approximately 1/10th in lifting scheme as compare to HE method) , PSNR (almost doubled in Lifting scheme as compare to HE method) are obtained for different images to show better enhancement using lifting based DWT method .
Keywords: Histogram Equalization (HE), Discrete Wavelet Transform (DWT), Mean Square Error (MSE), Peal Signal to Noise Ratio (PSNR) , Lifting scheme Forward Discrete Wavelet Transform (FDWT, Inverse Discrete Wavelet Transform (IDWT)
Scope of the Article: Image Processing and Pattern Recognition