An improved cultural algorithm and its application in image matching

Abstract

Cultural Algorithm (CA) are a class of computational models derived from observing the cultural evolution process in nature and is used to solve complex calculations of the new global optimization search algorithms. Aiming at the traditional cultural algorithm has poor precision and trap into local optimum of global optimization. In this paper, introduce the isolation niche technology into the traditional cultural algorithm. With improvements, the algorithm is less likely to trap in local optimum. According to the test of one set of benchmark function, the proposed algorithm has greater improvements than ordinal cultural algorithm in the aspects of convergence precision and stability. In this paper, introduce the proposed algorithm into the image matching problem, and the simulation test shows that the algorithm for image matching problem has made great effects in stability and convergence precision.