Vector-based image compression by current meta-heuristic algorithms

With the advancement of technology today, the quality of images created is increasing. Along with the quality, the sizes of these images and the time it takes to upload/download them are also increasing. Therefore, more storage and time are needed in this situation. Various image compression techniques are applied to images as a solution to these problems. Image compression techniques are divided into two categories as lossless image compression and lossy image compression. In lossless image compression, the image quality remains as unchanged as possible, but the compression ratio is low. In lossy image compression, more compression can be achieved by changing some pixels in the image, but the quality of the image will be lower than the original. Meta-heuristic algorithms have been developed based on the vital activities of any living creature in nature, expressing these activities mathematically. Activities such as hunting, exploration, chasing, and observation can be given as examples of these vital activities. In this study, the Fruit Fly Optimization Algorithm, inspired by the food search of fruit flies, the Firefly Optimization Algorithm, inspired by the mate selection of fireflies, the Particle Swarm Optimization Algorithm, based on the flight movements of birds, and the Bat Optimization Algorithm, inspired by the hunting movements of bats, have been examined. With the support of these algorithms, which continue to be used currently, lossy image compression has been applied to grayscale images that are widely used in the literature and have various contrasts. The performances of each algorithm on images with various contrasts have been examined, and it has been investigated which algorithm gives better results compared to other algorithms.