The Recovery Ratio Of The Interpolated Images And Wavelet Analysis

With the rapid development of computer technology, multimedia application is quite extensive. Especially image, it contains large amounts of data, requiring bigger storage space, but also to be wider transmission channel. Because the image exists similarities between the layers, and the same layer that exists information redundancy between adjacent points, which makes compressed images possible. In order to meet the application requirements of different users, not only to have a highe peak signal to noise ratio in the compression system, but also more new features.Wavelet transform can replace Fourier transform. It has multi-resolution analysis, and in the time domain and the frequency domain present the local characteristics of signal. In the field of static image wavelet coding, the most classic schemes are the embedded wavelet zerotree coding (EZW, Embedded Zerotree Wavelet Algorithm), and the resulting improvement of the SPIHT (Set Partitioning in Hierarchical Trees Algorithm, set partitioning in hierarchical trees algorithm) algorithm. The wavelet image coding techniques are widely used.This paper researches color image, first the image interpolation, image interpolation conbining with wavelet transform, then image restoration matching, finally image image compression encoding, the main results as follows:(1) Several common image interpolation algorithms are used in color images. Comparing and analysing theirs shortcomings, which designs an image restoration matching based on wavelet transform and the bilinear interpolation. In the same conditions, the new program has more superior effect, including the peak signal to noise ratio has improved greatly, and the efficiency has improved, so the reconstructed image meets the human vision.(2) The traditional wavelet transform is in the real domain. Even the initial image is an integer data, the wavelet coefficients are still real data. Now, constructing an "integer into integer wavelet transform", mapping integer data into integer coefficients of the wavelet, and that process must be reversible. Then by a discussion, integer wavelet conbines with the SPIHT algorithm, setting a SPIHT compression algorithm based on lifting wavelet.(3) In order to better achieve the effect of image compression, the SPIHT compression algorithm based on lifting wavelet is applied to the wavelet interpolation matching image of the above-mentioned program generated. Experimental results show, compared with the tradition SPIHT algorithm, new scheme has high availability, better compression effect. Especially at low bit rate, the compression effect has more obviously improved.