Hardware Implementation of Configurable Multi Image Fusion

Abstract

Multifocus images make it difficult to describe a scene accurately. Hence, the fusion of two or more images forms a good solution in order to obtain a new image containing more appropriate description of the scene than any of the input images. This paper presents a method of image fusion which uses stationary wavelet transform and fuzzy logic. By overcoming the disadvantages of discrete wavelet transform such as downsampling and shift variance, SWT gives a good quality image after fusion. First, the source multifocus images are decomposed using SWT and depending on appropriate fuzzy rule, images are combined to get better quality image. FPGA implementation of the system on Xilinx Virtex5 forms very effective method for image fusion of multifocus images as it uses less FPGA resources like number of slice registers, number of slice LUTs, number of block RAM and number of bonded IOBs.

Keywords: Stationary wavelet transform (SWT), Xilinx system generator, FPGA, fuzzy logic