Illumination-Invariant Image Matching Based on Simulated Kalman Filter (SKF) Algorithm

Abstract

In this paper, a novel image template matching approach to tackle illumination-invariant problem has been proposed. The traditional algorithm to solve image matching problem takes a lot of memory and computational time. Therefore, the image matching problem is assigned to optimization problem and can be solved precisely. Although there are some methods presented recently for image matching illumination-invariant, all of them have limitations in term of dealing with the pixels complexity and many unknown parameters in a certain algorithm. The search of the image template has been performed exhaustively by using Simulated Kalman Filter (SKF) algorithm. The experiment is conducted using an image taken from the database and the contrast image is changed to get the illumination effect. Experimental results show the comparison between SKF and Particle Swarm Optimization (PSO) to get the performance of the correct matching. The percentage of the matching result for the image within six conditions are 24%, 16%, 16%, 12%, 28% and 4% accordingly, which are higher than the PSO algorithm, which obtained 0% correct matching for all conditions.