Research On JPEG Image Compression Algorithm Based On Wavelet Transform

Wavelet transform has time and frequency domain characteristics of the dual locality, which is beneficial to analyze the time-frequency characteristics of signals. Therefore, the wavelet transform has been widely studied and applied in image compression field. Comparing to the traditional image compression based on wavelet transform has the better compression efficiency, and the ability to achieve progressive transmission of compressed signals. The multi-resolution analysis characteristic of wavelet transform provides an advanced mechanism for adapting human visual feature, and the wavelet transform image data maintains the original image fine structure in various resolutions, in order to remove the other forms of redundancy information in image conveniently.In this paper, on the basis of studying image compression standard JPEG and JPEG2000, systematically analysising of the decomposition and reconstruction of wavelet image, focusing on research of image compression algorithm based on the discrete cosine transform and wavelet transform. Simulation experiments using MATLAB7.0 and then testing and analysising of the images shows that wavelet compression algorithms has excellent compression ratio, high compression efficiency and low energy loss.Finally, researching the selection of optimal wavelet bases and discussing three selected factors affected image compression, and selecting five different wavelet functions for image compression, comparing data from compression, ultimately reaching a conclusion that the better compressed consequent can be obtained by using Dmeyer wavelet function.