Noise-refined image enhancement using multi-objective optimisation

This study presents a novel scheme for the enhancement of images using stochastic resonance (SR) noise. In this scheme, a suitable dose of noise is added to the lower quality images such that the performance of a sub-optimal image enhancer is improved without altering its parameters. Image enhancement is modelled as a constrained multi-objective optimisation (MOO) problem, with similarity and some desired image-enhancement characteristic being the two objective functions. The principle of SR noise-refined image enhancement is analysed, and an image-enhancement system is developed. A genetic algorithm-based MOO technique is employed to find the optimum parameters of the SR noise distribution. Several image-enhancement examples are provided, where the efficiency of the presented method in several real-world applications is shown.