Search

Aiming at the discontinuity of the traditional hard threshold function at the threshold and the constant deviation between the original wavelet coefficient and the wavelet estimation coefficient in the soft threshold function, this paper proposed an image denoising algorithm based on the improved threshold function. This algorithm combined the advantages of the improved threshold function, dynamically selected the fixed threshold by setting appropriate adjustment parameters, and added the adjustment factors to reduce the constant deviation between the original wavelet coefficient and the estimated wavelet coefficient, there by improving the degree of approximation of the reconstructed image and the original image. The improved threshold function satisfied continuity at the threshold while satisfying the asymptotic and higher order conductibility of the function. Simulation results show that the visual effect is good, while using the improved threshold function for image denoising. In comparison, both PSNR and SNR are improved,MSE is reduced, and the denoising effect is optimized . [ABSTRACT FROM AUTHOR]

Aiming at the problems of selecting the adjustment parameters in threshold and threshold function, this paper proposed a signal denoising method based on artificial bee optimization algorithm and new threshold function. Firstly, it carried out the theoretical analysis of the new threshold function to verify its continuity, high-order differentiability and parameter adjustability. Secondly, according to the minimum mean square error (MSE) strategy, it used the artificial bee colony optimization algorithm to optimize the thresholds and adjustment parameters of each decomposition layer, and then obtained the optimal denoised signal. Finally, it carried out the simulation experiment to verify the denoising effect according to the signal-to-noise ratio (SNR) and MSE. The experimental result shows that the threshold parameters selected by artificial bee optimization algorithm and the new wavelet threshold function can effectively denoise a noisy signal. [ABSTRACT FROM AUTHOR]

Aiming at the limitations of methods on speech enhancement by traditional threshold methods in wavelet domain, this paper proposed a new speech enhancement algorithm based on the tunable Q-factor wavelet transform and separation of voiced signal and unvoiced signal. Firstly, it separated voiced signal and unvoiced signal with zero-crossing ratio and short time energy. The adaptive threshold values combined the energy of coefficients and the variance of noise in different scales, respectively. Then it applied the improved Donoho threshold value and threshold function to process wavelet coefficients of voiced signal and unvoiced signal, and estimated the original coefficients from noisy coefficients. Lastly, it used the inverse transform to obtain the original speech signal which the noise was removed. Comparing with the other current classical algorithms, experimental results show that the modified algorithm improves the effect of de-noising and speech intelligibility in white Gaussian noise and colored noise background. [ABSTRACT FROM AUTHOR]