Genetic Algorithm Based Parameter Identification of Defocused Image

A method to identify the degradation parameter of the defocused blur image based on genetic algorithm is proposed. In this technique, a FFT transform is performed in the frequency domain to estimate the range of defocus parameter, and then genetic algorithm is used to search in this domain to find the accurate value of the radius. We experimentally illustrate its performance, Results show that the proposed  parameter estimation technique is closely to the real defocus radius and more robust to noise.