An Image Denoising Algorithm Based on Improved Wavelet Threshold Function

Abstract

In the field of image denoising research, the technique of wavelet threshold denoising has been widely used. Aiming at the shortcomings of traditional hard threshold and soft threshold denoising, an improved threshold function is proposed for image denoising in this paper. Two tuning parameters are added to this threshold function to improve the flexibility of the function. In the evaluation of denoising performance, this paper uses the peak signal to noise ratio (PSNR) and mean square error (MSE) as evaluation indicators. Experimental results on Boats images show that algorithm proposed in this paper improves the PSNR by 0.1 dB and 0.12 dB and reduces the MSE by 2.35% and 2.81%, respectively, compared with the algorithms in reference [6] and reference [7]. The experimental results on other images also show that the algorithm proposed in this paper also has some improvement in evaluation indexes compared with several comparative algorithms.