Wavelet basis selection based on fractal dimension in image compression
Tang Guowe · Journal of Northeast Petroleum University · 2014
For the variety of wavelet function,the quality of the reconstructed image compressed by using different wavelet basis is different.In order to select a suitable wavelet basis,a fractal dimension based wavelet basis selection method is proposed.By calculating the fractal dimension of the relevant images with differential box-counting method,the images are divided into different categories according to the fractal dimension value.A representative image for each category is chosen,and SPECK algorithm is used to carry out compression operation under different wavelet bases.Then according to the peak signal to noise ratio of the reconstructed image,a suitable wavelet basis is determined for each image category. Example analysis demonstrate that the method improves the PSNR of wavelet based image coding algorithm.