Selection of Wavelet from Wavelet Families to Facilitate the evolution of Color Image Denoising
Reena Thakur, Rishu Gupta, Supriya Shukla · 2013
Denoising of image is very important and inverse problem of image processing which is useful in the areas of image mining, image segmentation, pattern recognition and an important preprocessing technique to remove the noise from the naturally corrupted image by the different types of noises. The different wavelet families are one among the diverse methods for recovering infinite dimensional objects like curves, densities, images etc. The wavelet techniques are very effective to remove the noise also because of its capability to confine the power of a signal in little convert of energy values. This paper reviews on the existing various wavelets by using multiplicative and additional noise models which includes Salt and Pepper noise, Gaussian noise, Speckle noise. Further, it analyses, examines and compares various wavelet function families like Haar, Symlets, Coiflets, Daubechies, Meyer and Biorthogonal using different testing images. The experimental results shown to be precised in terms of SNR and varience in noise. The results consider the quantitative measures of comparing the denoised images as output of different wavelets based on the hard and soft thresholding and one level or two level decomposition.