| The21st century is an informationization century, and the digital image plays an important role in the world of information. With the development of society and science and technology, all kinds of digital images have been an integral part of our lives and our work. The imaging technology of digital camera develops rapidly in that the higher digital image definition is needed in our daily life, which brings more enjoyable and convenient lives to our life and work. However, the enhancement of clarity of digital image causes the quick increasing of image sizes. The large data of digital image has had a great impact on the stress on the local image storage and remote image transmission of multimedia applications, which not only is a huge amount of storage required, but also the bandwidth limitations of communication channel makes it hard for digital image to accomplish fast transmission, browsing and data sharing online.Obviously, there are three ways to solve the contradiction among the large image data, communication bandwidth and storage limitation in practical using:increase channel band, storage capacity and apply the proper image compression. The first two ways can help to solve the problem but considering the development of digital imaging technology and the size of single picture increasing, the work on these two aspects are always in vain. Moreover, we can not gain the increasing bandwidth and higher storage capacity, which means these two ways are a temporary palliative. The key to solve the problem is to get the images compression.Due to the urgent need of high rate of compression image algorithm, the image compression has become a hot area of research for more and more researchers and many more excellent and effective algorithm has been coming out. In recent decades, because of the merit of time-frequency localization of wavelet transform, it is widely applied in the area of digital image compression. As to the research and application of wavelet transform image compression are not only interactive in academic realm but also in business. Some foreign companies apply this technology on the internet such as image data transmission to provide a commercial service, which promotes to relieve the shortage of network bandwidth and speeds up the image information transmission. As an excellent image compression algorithm, the wavelet transform is of wide potential applications in the respect of this area and plays a significant role, which has greatly improved the technology’s promotion and application.The less important data coming from the image wavelet transform, which need to be quantified and encoded in order to be removed for realization of digital image compression. Currently zero tree encoding is a common coding technique of image compression algorithm. Based on its rules, Shapiro and some other researchers present the embedded image coding using Zero tree of Wavelets coefficients, EZW. It is the first way to do wavelet image compression algorithm by using zero tree encoding, which take advantage of similarity of sub-bands in wavelet image to sort the coefficients according to their significance, to quantify the coefficients and to get bitstream from coefficients significance order. The EZW adopting the embedded coding technique makes it happen that image can be progressive transformed on the internet and the users can make sure any compression rate as needed. Therefore, EZW gets a lot of attention after its presentation.Firstly, this paper simply introduces the current situation of digital image compression and its necessity and feasibility. Then, the related basic knowledge about the digital image compression is introduced in detailed. What’s more, the basic theory of wavelet transform is presented. At last, the application of wavelet transform in the image compression area gets to be further description, especially about the EZW. At the same chapter, addition to its theory, the merits and the shortages are also concluded. By analyzing its advantages and disadvantages, this paper is going to present a improved algorithm.Based on the experimental data, we evaluate the reconstructed image through subject assessment and objective metric.From subjective assessment, we found out that the quality of reconstructed image with the reformed EZW is better than the original EZW. According to objective metric, at the same coding rate, the reformed EZW makes more improvement of peak value Signal-to-Noise Ratio(PSNR) of reconstructed image. That is to say, no matter the subjective or objective standards, it is efficient and feasible to make the EZW compression algorithm reformed. |
