Research On Still Image Compression Coding Based On Wavelet Transform

As the wavelet transform has good locality characteristics both in time-plane and frequency-plane, so wavelet transform has been widely used in image compression in recent years. Those image coding devices that based on wavelet transform can obtain higher efficiency in compression, and facilitate to construct an embedded bit stream. Embedded image coding devices have many advantages, such as both airspace and quality have scalability. So, they are ideal for many applications, such as network transmission, wireless communications, telemedicine, digital libraries, etc. In this paper, deeply study of the characters of the wavelet transform coefficients, then combine with EZW and SPIHT, respectively propose two improved algorithms.In this paper, an overview of innovative work:(1) The proposed algorithm can be viewed as improved version of EZW algorithm, where the lowest frequency subband is coded separately from other highpass subbands and it pay more attention to the edge of one image because human visual system is sensitive to edge information. Experimental results show that the new approach performs better than EZW in the recovery image quality , especially in the case of low rate.(2)SPIHT is computationally very fast and among the best image compression algorithms known today. It can encode effectively, but without arithmetic coding, so, there will be have some redundancy. This paper proposes a new method that put SPIHT combines with arithmetic coding based on analyzing the output data stream of SPIHT encoder. A large number of experimental results show that this method saves a lot of bits in transmission, further enhanced the compression performance of SPIHT algorithm.