作者: Ming Yin , Wei Liu , Jun Shui , Jiangmin Wu
DOI: 10.1155/2012/493976
关键词: Artificial intelligence 、 Second-generation wavelet transform 、 Wavelet transform 、 Mathematics 、 Lifting scheme 、 Stationary wavelet transform 、 Wavelet 、 Wavelet packet decomposition 、 Discrete wavelet transform 、 Mathematical analysis 、 Pattern recognition 、 Harmonic wavelet transform
摘要: The quaternion wavelet transform is a new multiscale analysis tool. Firstly, this paper studies the standard orthogonal basis of scale space and in spatial , proves presents wavelet’s function concepts structure. Finally, applied to image denoising, generalized Gauss distribution used model QWT coefficients’ magnitude distribution, under Bayesian theory framework, recover original coefficients from noisy coefficients, so as achieve aim denoising. Experimental results show that our method not only better than many current denoising methods peak signal noise ratio (PSNR), but also obtained visual effect.