Genetic algorithm-based parameter selection approach to single image defogging

• Two main sensitive parameters are distinguished from other less sensitive ones.
• The two sensitive parameters are automatically determined using a genetic algorithm.
• A proper defogging effect assessment index is used as the fitness function of the genetic algorithm.

Image defogging is widely used in many outdoor working systems. However, owing to the lack of enough information to solve the equation of image degradation model, existing restoration methods generally introduce some parameters and set these values fixed. Inappropriate parameter setting will lead to difficulty in obtaining the best defogging results for different input foggy images. This letter proposes a novel defogging parameter value selection algorithm based on genetic algorithm (GA). We mainly focus on the way to select optimal parameter values for image defogging. The proposed method is applied to two representative defogging algorithms by selecting the two main parameters and optimizing them using the genetic algorithm. An assessment index of image defogging effect is used in the proposed method as the fitness function of the genetic algorithm. Thus, these parameters may be adaptively and automatically adjusted for the defogging algorithms. A comparative study and qualitative evaluation demonstrate that the better quality results are obtained by using the proposed method.