Image Denoising Based Using Hybrid Techniques Mixed between (Hard and Soft Threshold) With Multiwavelet Transform and Multi-Stage Vector Quantization


The Image denoising naturally corrupted by noise is a classical problem in the field of signal or image processing. Denoising of a natural images corrupted by Gaussian noise using new techniques of multiwavelet techniques depended on losing of some error are occurs in reconstruct in pre-processing of multiwavelet then the new multiwavelet are very effective because of its ability to reduce losing of some data in reconstruct. Multi-wavelet can satisfy with symmetry and asymmetry which are very important characteristics in signal processing The better denoising result depends on the degree of the noise. Generally, its energy is distributed over low frequency band while both its noise and details are distributed over high frequency band. Corresponding hard threshold used in different scale high frequency sub-bands. In this paper proposed to indicate the suitability of different multiwavelet based on using the mixing between Hard and Soft threshold that named as Hybrid threshold technical depended on the parts of the multiwavelet decomposition, according the value of noise in the decomposition parts used the threshold techniques for example using the soft threshold on the two first parts LL and LH decompositions because that the amount value of pixels in this part is Low frequency and some of Hard and then using the Hard threshold of the remaining two parts HL and HH part because the amount value of pixels is High frequency, then the performance calculation of image denoising algorithm in terms of PSNR value. Finally it's compare between multiwavelet traditional techniques Hard, Soft threshold and produced best denoised image using (Hybrid threshold) image denoising algorithm in terms of PSNR Values based on mixed thresolding (hard and soft thresolding) by using the first (Hard threshold) in LL and LH part and the second (soft thresolding) in HL and HH part from multiwavelet decomposition.