Improved adolescent identity search algorithm for block-based image compression

Image compression has become an indispensable part of the digital world, reducing the costs of storing and transmitting high-quality images. Vector quantization method, which stands out among image compression methods, facilitates the storage and transmission of images without compromising image quality in the process of reducing their size. It uses a codebook to represent similar patterns in a set of images in this method. If the optimum codebook design is achieved, the loss in image quality is minimized by increasing the data compression ratio. It is efficient to use nature-inspired metaheuristic algorithms in the optimization of complex and multidimensional problems such as optimum codebook generation in the image compression process. In this paper, Adolescent Identity Search Algorithm (AISA), which simulates the identity search process of adolescents, is designed for the first time of optimum codebook generation in the block-based image compression process. In addition, improved AISA with various modifications has been proposed to avoid the capture of the conventional AISA by local minimums and to customize the search space. The proposed improved AISA is compared with various well-known metaheuristic algorithms in the literature, and the visual and numerical results are evaluated on various metrics. Based on the simulation results, it has been shown that the proposed IAISA is more efficient than the other metaheuristic algorithms for block-based image compression.