关键词: 乘性噪声, 核回归, 全变差, 自适应, 去噪

Abstract: To remove the multiplicative noise better, a new three-stage method for multiplicative noise removal was proposed. In the first stage, log-image was processed by adaptive Steer Kernel Regression (SKR). Then in the second stage, the Total Variation (TV) regularization method was used to amend the image obtained. At last, via an exponential function and bias correction, the result was transformed back from the log-domain to the real one. The new method combined the advantages of steer kernel regression and total variation method. The experimental results show that the new method is more effective to filter out multiplicative noise.

Key words: multiplicative noise, kernel regression, Total Variation (TV), adaptive, denoise