Integration of Multispectral Images Using Two-Dimensional Wavelet Transformation
AL-Rafidain Journal of Computer Sciences and Mathematics,
2010, Volume 7, Issue 2, Pages 89-99
AbstractImage fusion is an effective tool to integrate multi-source image, where the purpose of the process access to information integration concept, and make the resulting image more suited to human vision, as well as digital processing. This integration leads to a technology that joins the resources in order to increase and clarify the information that serve the visual and automatic interpretation.
The current research aims to use wavelet transformation with two dimensions (2D-DWT) in the integration of multispectral images (images taken by visible band) and others taken by IR (Infrared band) to obtain a single and clear image, and shows in accurate details of the original images.
This algorithm programmed by MATLAB tools (MATLAB R2008a) is applied to the two images, the dimensions of each image 512×512 pixels. Where the fusion process depends on two types of mathematical models, the Maximum, and Average. High pass Combination and Low pass Combination filters were used. Using filters db4, coif4 and sym4 and then the Entropies of images were calculated to measure the amount of information in the resulting image.
This paper shows that the 2D-DWT is used to integrate images using the image fusion ,and it gives good results in increasing the amount of generated information, increase the efficiency of the image , increase reliability , and reduce the uncertainty-surrounding some of the features in the images which are not clear.
- Article View: 75
- PDF Download: 85