Color image quantization using the shuffled-frog leaping algorithm

Abstract

[EN] Swarm-based algorithms define a family of methods that consider a population of very simple individuals that cooperate to solve a difficult problem. The Shuffled frog-leaping algorithm is a method of this type that has been applied to solve different types of problems. This article describes the application of this algorithm to the color quantization problem. Although the selected method was developed to solve optimization problems, this work shows how it can be adapted to solve the problem proposed. The proposed method uses the mean squared error as the objective function of the optimization problem to be solved. To reduce the execution time of the algorithm, it is applied to a subset of pixels of the original image. As a result, a quantized palette is obtained that is used to define the quantized image. Computational results indicate that the proposed method can generate a quantized image with low computational cost. Moreover, the quality of the image generated is better than that of the images obtained by several well-known color quantization methods.