The Second Generation Wavelet-based Image Compression Coding

Image information is an important headspring for people to know the world and ourselves. But the dataquantity of image is huge, which not only requires high storing and operating capacity for computers, but alsorestricts the channel transfer rate of image communication. Therefore, in order to store, process and transferthe data, compress processing must be done for image information. The traditional image compresses of adopting the DCT transformation to code could appear "a block effect", the efficiency compress at the same time is also not high,a wavelet image coding is in recent years along with a development and analyzes of thewavelet theories to put forward a kind of way with very good development foreground of image code the method .the wavelet analysis have many advantages of time-frequency analysis, many differences analysis etc., easily combining together with the person's eye sense of vision characteristic, can get a good compression result, it is specially and suitably suit for compressing process of this signal, so the method of a wavelet transformation is high among the people value, and appearing various image coding methods according to a wavelet transformation . This paper discusses it in such condition.This paper studies to code according to the image of a transformation of lifting.Studied to rising the method, zero trees to code the calculate way in a specific way .Rising the method belongs to the new method of the wavelet of the next generation structure, since inherits the first generation characteristic of the wavelet, and then has not to be depend on Fourier transformation, at the same time it's easy to structure integral characteristic of the wavelet .This paper elaborates and discusses towards rising principle and key technique in which of method to make detailedly, chooses Haar wavelet of lifting methods of realization lifting wavelet and from integral to integral, for the behind further research coefficient codes to work well the preparation.Embedded image coding using zerotrees of wavelet coefficients (EZW) algorithm and Set partitioning in hierarchical trees algorithm (SPIHT) are researched and simulated. On the basis of it, the disadvantages of the algorithms are analyzed andan improved algorithm based on HVS------which reserves low frequency components.The simulation of the algorithm is realized, and the results show that the algorithm can improve the peak signal noise ratio (PSNR) and reduce the mean square enor (MSE) compared with EZW. Last ,using VC program the algorithm. Test images, it confirms that that method can carry out the image transformation from completely compression of lossless to the compression of losing comparing with the method according to a wavelet transformation and zero amount of trees measure , it operates more simply and the speed is quicker, the composition image has quantity higher, obtains the decent result.