Noise removal is one of the important steps in digital image processing. In this paper, we have proposed and implemented a new method based on segmentation and two levels of decomposition TWT (Toggle wavelet transform) in which noise will be removed and at the same time edges will also be preserved by making use of standardization formula. Nano tree region based segmentation will divide the image into homogeneous region and preserve the edges. It will divide the image into nine regions and proceed further.Toggle wavelet transformation will compute the average and difference between the alternate rows and columns of matrix and stores the details and reconstruct the image by inverse transformation. The image is divided into two images with horizontal approximation and horizontal detail image by applying filter banks such as low pass and high pass filters on rows. Further each image is divided into two like 20, 21, 22, 23 etc. At the end of the paper, we can see the denoised images (approximation, vertical, horizontal, diagonal) based on segmentation and TWT which preserves edges. The paper also lists the time taken to execute each image.
Volume 11 | 02-Special Issue
Pages: 178-184