By Topic

Quantitative effects of discrete wavelet transforms and wavelet packets on aerial digital image denoising

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Zhengmao Ye ; College of Engineering, Southern University, Baton Rouge, LA 70813, USA ; Habib Mohamadian ; Yongmao Ye

The objective of image denoising is to remove the noises and to retain important image features as much as possible. Linear approaches could be effective for some simple cases with slowly varying noises, but not for other slowly varying noise cases and rapidly varying noise cases. As a nonlinear wavelet based technique, the wavelet thresholding is effective to denoise blurring aerial images. Either the discrete wavelet transform or wavelet packets technique can be employed using wavelet decomposition. At each level of wavelet decompositions, the digital image is split into four subbands, representing approximation (low frequency feature) and three details (high frequency features) in horizontal, vertical and diagonal directions. The proposed soft thresholding wavelet decomposition at multiple levels is a simple and efficient method for reduction of noises. For multiple level decompositions in terms of both the discrete wavelet transform and wavelet packets techniques, the approximation component will always be decomposed at each level. If the detail components are further decomposed as well similar to that of the approximation, it is the wavelet packet approach, otherwise it is the discrete wavelet transform. On a basis of the proposed thresholding technique at different levels for wavelet denoising, objective metrics can be introduced also to evaluate and compare the denoising effects of the discrete wavelet transform and wavelet packets quantitatively rather than qualitative observation, such as the metrics of the discrete entropy, energy and mutual information.

Published in:

Electrical Engineering, Computing Science and Automatic Control,CCE,2009 6th International Conference on

Date of Conference:

10-13 Jan. 2009