Multi-modal Medical Image Fusion Based In Multiscale Alternative Approach To Cross Bilateral Filter

Résumé: Image fusion is a process of combining additional information from source images into a single image. We are interested in this paper in the medical field where image fusion is a great advance in medical imaging because it helps the doctor to make his diagnosis. The aim of this paper is to propose a multi-scale method based on the cross bilateral filter, visual salience detection, and weight map construction for image fusion. In the proposed image fusion approach, the cross bilateral filter is applied to one source image to obtain a base layer by considering another image for the guiding purpose. Whereas, the detail layers are computed by finding the difference between source image and the base layers and used for visual saliency extraction and weight maps construction . Later, base layers are averaged to get the final base layer. Finally, final base and detail layers are combined to obtain the fused image. The proposed method has been tested on a set of gray scale and color medical images. The results show that the proposed method can be competitive or even outperform the methods in comparison in terms of subjective and objective evaluations.

Publié dans la revue: المجلة الجزائرية للعلوم