An Approach to Study Image Denoising Using Doubly Sparse Transform Technique
Published: 2016
Author(s) Name: Laxmi Priya Sahu, B. N. Biswal |
Author(s) Affiliation: Asst Prof,Electronics & Tele-Communication Engg,Bhubaneswar Engg College,Bhubaneshwar,Odisha,India
Locked
Subscribed
Available for All
Abstract
In this Paper the sparse domain of signals in a certain area or dictionary has been widely used in many applications in image, audio, biological and other signal analysis. Analytical Sparse transforms such as discrete cosine transform (DCT) and its counterpart i.e. wavelet transform (WT) have been extensively used in the areas of image compression standards where as synthesis sparsifying dictionaries have become extensively used especially in applications such as image de-noising and medical image reconstruction. In this work, we discuss about the square sparsifying transforms which is the product from a fixed, fast transform so as to consider the DCT and an adaptive constrained matrix to be sparse. Such transforms can be studied and implemented efficiently.
Keywords: Dictionary Learning, Sparse Representation, Image de-noising, Wavelet Transform, Discrete Cosine Transform
View PDF