International Journal of Computer Science Issues

More than a traditional journal...

This paper introduces the concept of image fusion technique for impulse noise reduction in digital images. Image fusion is the process of combining two or more images into a single image while retaining the important features of each image. Multiple image fusion is an important technique used in military, remote sensing and medical applications. The images captured by two different sensors undergo filtering using vector median or spatial median filter based on the noise density in the image. The filtered images are fused into a single image, which combines the uncorrupted pixels from each one of the filtered image. The fusion algorithm is based on Bi-dimensional Empirical Mode Decomposition (BEMD), which decomposes an image into residue and IMF components. Different fusion rules are used to combine IMFs and Residual components. Finally, the image is recovered using inverse BEMD. The performance evaluation of the fusion algorithm is evaluated using structural similarity index (SSIM) between original and fused image. Experimental results show that this fusion algorithm produce a high quality image than individually filtered image.