论文部分内容阅读
Based on the Markov random field (MRF) theory, a new nonlinear operator isdefined according to the statistical information in the image, and the corresponding 2Dnonlinear wavelet transform is also provided. It is proved that many detail coefficientsbeing zero (or almost zero) in the smooth gray-level variation areas can be achievedunder the conditional probability density function in MRF model, which shows that thisoperator is suitable for the task of image compression, especially for lossless codingapplications. Experimental results using several test images indicate good performancesof the proposed method with the smaller entropy for the compound and smooth medicalimages with respect to the other nonlinear transform methods based on median andmorphological operator and some well-known linear lifting wavelet transform methods(5/3, 9/7, and S+P).
Based on the Markov random field (MRF) theory, a new nonlinear operator isdefined according to the statistical information in the image, and the corresponding 2Dnonlinear wavelet transform is also provided. It is proved that many details coefficients being zero (or almost zero) in the smooth gray-level variation areas can be achieved by the conditional probability density function in MRF model, which shows that this opener is suitable for the task of image compression, especially for lossless coding applications. smaller entropy for the compound and smooth medical images with respect to the other nonlinear transform methods based on median and morphological operators and some well-known linear lifting wavelet transform methods (5/3, 9/7, and S + P).