Image Denoising, Deblurring and Object Tracking: A New Generation wavelet based approach

by Nguyễn Thanh Bình

Noise reduction is an important step for development of any sophisticated algorithms in computer vision and image processing. A tradeoff between the removed noise and the blur in the image always exists. The capability of the wavelets to give detail spatial-frequency information is the main reason for the use of wavelets. This property promises a possibility for better discrimination between the noise and the real data. Successful exploitation of the wavelet transform might reduce the blurring effect or even overcome it completely. Object tracking is a problem of estimating the positions and other relevant information of moving objects in the sequences of image video. The main difficulties in reliable tracking of moving objects rapid appearance changes caused by image noise and interaction between multiple objects. In a long sequence of image video, if the dynamics of the moving object is known, prediction can be made about the positions of the object in a particular frame. This information can be combined with the actual image observation to achieve more robust results. We have explored a possibility to use wavelet transform for object tracking.

148 pages, Paperback

Published May 18, 2013