Image Denoising with Adaptive Threshold in Wavelet-Domain
Jiuxian Li · Signal Processing · 2007
Donoho's denoising scheme by thresholding the wavelet coefficients kills too many wavelet coefficients without consid- ering the image's local characteristic.We introduce an adaptive wavelet coefficients shrinkage method,which can modify the magnitude of shrinkage by considering neighboring wavelet coefficients,variance of noise and decomposition levels.Therefore,we can use more wavelet decomposition levels than other wavelet shrinkage methods to recover the noisy images.We also discussed how to preset the num- ber of decomposition level,window size and the complexity of the algorithm.The proposed method outperforms the other methods in the literatures,while its implement and concept are both simplicity.The results have good visual effect and low mean square error.