A NOVEL PROPOSED FUSION METHOD

Abstract

In this paper, two proposed fusion methods will be proposed for merging important information from more than one source for the purpose of either making final image with high quality compared to the input images or making final image holding more information that captured in the input images depending on some important extracted features from the input images. The first method merge images depending on edges information in each pixel by applying one order prewitt method for edge detection then calculating the ratio related to each image for this position then calculating new pixel value for this place in the output image, while the second proposed method working by estimating texture unit number value in each pixel in the input images and merging pixels by applying proposed techniques for the purpose of calculating output pixel which reflect good degree of the proposed feature with accepted compatible degree with neighbouring eight pixels. The results of the proposed methods checked with some suitable metrics for measuring the quality of the output images compared with the standard methods like maximum, minimum, ratio, wavelet,… and the result of the proposed method reflect good and perfect quality for the output images and this result can be expected because fusion operation performed according to effected visible features that making fused images with good quality.