Mixed noise reduction method based on fuzzy morphological filtering

To remove mixed noises from the image, we studied the noise reduction performance of fuzzy morphological filter and introduce a new method to reduce noise. Firstly, the method uses S-function to fuzz an image, and then do fuzzy morphological opening-closing operation with the fuzzed image. To improve the noise reduction performance, we adopt double structural elements which combined fuzzy square structural elements with fuzzy linear structural elements. Experiment show that the proposed method may effectively reduce impulse noise, Gaussian noise, and mixed noise. Especially, it has obvious advantages in dealing with the image that has serious impulse noise.