A kind of effective method of removing compound noise in image
Hongyan Chen · 2016
The noise is generally not a single type in digital image, which is usually composed of Gaussian noise and pulse noise. So the signal to noise ratio of the image is low and the noise in images needs to be processed. The traditional denoising algorithm and various improved algorithm is effective for single Gaussian noise or impulse noise mostly, these denoising algorithms are unable to obtain satisfactory denoising effect in dealing with the mixed noise. Some existing hybrid denoising methods usually change distribution characteristics of another noise when removing one noise at the same time, so subsequent denoising effect is affected. To resolve the problem, a composite image denoising method based on the combination of independent component analysis and traditional spatial denoising method is proposed. Simulation experiment results show that the method can effectively preserve edge details and other information of image when removing mixed noise.