HRFA Based Image Denoising With Edge Preserve Segmentation

Apply

HRFA Based Image Denoising With Edge Preserve Segmentation

Journal Title: IOSR Journal of Electronics and Communication Engineering(IOSR-JECE) - Year 2018, Vol 13, Issue 6

Abstract

Denoising of images is an emerging technique in the present day research works, since it is used in the various processing in image. Considerably, the denoising of image is a method in which the image is restored by removing the unwanted distortions or noises from the image in order to obtain a high visually enhanced image. But existing denoising algorithms mainly focus on the noise reduction instead of focusing on the edge preserving. Hence the finally obtained image is of low quality. Therefore in our paper we proposed to eliminate the noise such as zero mean white Gaussian noise from the image by using the high resolution frequency analysis. For this process we calculate the texture features of the image in order to implement the segmentation make possible in noisy image. Hence, we implement different segmentation techniques for different images corrupted with different noise in order to obtain the efficient segmented image. In our paper we implement mean shift clustering for segmentation of object from the image. In addition to this segmentation we also use edge segmentation technique with different parameters. And then the non uniform regions obtained from the segmentation process are analyzed based on the Mask NHA which is an extended version of 2D NHA. However, the calculation of the different parameter values takes more time and it is not an easy job. After studying and conducting experiments on the image by using Mask NHA method it is noticed that this method provides valid and better results compared to state of art methods such as BM3D and LRF.

Authors and Affiliations

K. M. Hemambaran, Dr. S. A. K. Jilani

Keywords

EP ID EP439330
DOI 10.9790/2834-1306012734.
Views 129
Downloads 0