Research On The De-noising Algorithm Of Underwater Image Based On Wavelet

Abstract:In the development and exploration of ocean, underwater image plays an important role, and is an important means to study marine resources. Because of the complexity of the underwater environment and poor imaging conditions, underwater image usually contains a variety of noise. In order to obtain a clear underwater image, underwater image de-noising is an important and challenging work. This paper focuses on the research of underwater image wavelet de-noising algorithm, and in order to improve image quality. The main research work and innovations are as follows:Proposes a median filter method based on local mean and variance for image preprocessing, and carries out an experiment. The results show that the method can remove the impulse noise in the image under water.According to the noise characteristics of underwater image, this paper proposes a mean of cyclic double threshold filtering of wavelet. The method calculate the difference of original image’s noise standard deviation and the obtained noise image’s standard deviation after wavelet transform, and determine the optimal threshold with the differentials. Then it sets the upper and lower thresholds with the optimal threshold and the estimated threshold of the original image and de-noise underwater image with the tow thresholds. And through the simulation experiment, the results show that the improved de-noising algorithm can get better de-noising effect.Finally, this paper realizes the modified de-noising algorithm of underwater image on the DM642image processing system platform, and de-noise for the acquisition of the underwater image, validate the feasibility and effectiveness of the method proposed in this paper.