The performance evaluation of the Cat and Particle Swarm Optimization Techniques in the image enhancement

The aim of this study is to adjust automatically the contrast level of the distorted images. Since the images are in two- dimensional and multiple parameters are need to be tuned in the enhancement techniques, the classical algorithms increase the computational complexity with less enhancement success. However, optimization algorithms are able to find the optimal solutions to different types of problems. In this respect, the most appropriate solution can be found faster and more effectively by using optimization algorithms in the enhancement techniques. In this study, two swarm based optimization technique, Particle Swarm Optimization (PSO) and Cat Swarm Optimization (CSO), are used for the parameter tuning process in the histogram stretching image enhancement technique. Techniques have been tested on a data set containing four level contrast distortions. Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) metrics are reported as the performance metrics. The results are examined and compared with respect to each other and to one of the parameter independent image enhancement technique, histogram equalization. Even though CSO and PSO are not guaranteed to succeed in each result, they have presented more successful results compared to traditional histogram equalization method when it comes to converging to optimal solutions.