Image De-Noising with the Aid of Dual Tree Wavelet Transform And Artificial Bee Colony Optimization Algorithm

Volume 3 Issue 4 October - December 2016

Research Paper

Image De-Noising with the Aid of Dual Tree Wavelet Transform And Artificial Bee Colony Optimization Algorithm

Yugandhar Dasari*, S.K. Nayak**

* Associate Professor, Department of Electronics and Communication Engineering, Aditya Institute of Technology and Management, Tekkali, Andhra Pradesh, India

** Professor, Department of Electronic Science, Berhampur University, Berhampur, Odisha.

Dasari, Y., and Nayak, S. K. (2016). Image De-Noising With The Aid Of Dual Tree Wavelet Transform And Artificial Bee Colony Optimization Algorithm. i-manager's Journal on Image Processing, 3(4), 9-18.

Abstract

Determination of the threshold value is extremely an important part in wavelet based image de-noising. Finding appropriate threshold value can be done either by using deterministic approaches or soft computing algorithms. Artificial Bee Colony (ABC) is one of the algorithms motivated by the intelligent behaviour of honey bees. In this paper, a nature inspired population based image de-noising technique has been implemented to find the dynamic threshold value using an ABC algorithm by using Dual Tree Complex Wavelet Transform (DT-CWT). The DT-CWT is a relatively recent enhancement to the Discrete Wavelet Transform (DWT) with two additional properties known as shift invariance and directional selectivity. These additional features are in turn used to preserve geometric image features like ridges and edges. The performance of the proposed method has been compared in terms of Peak Signal to Noise Ratio (PSNR) with conventional wavelet thresholding using DT-CWT and Adaptive Median Filter (AMF) techniques.