论文部分内容阅读
文章利用小波变换多分辨分解特点和Ｓｈａｐｉｒｏ提出的ＥＺＷ编码的优点，实现了一种基于零树结构的可伸缩性编码新方法，其编码特点是可以输出分辨率多位率压缩位流，有效地改进了ＥＺＷ方法只能应用于单一分辨率视频服务的不足。该算法首先对小波变换系数进行一致标量量化；其次对量化后不为零的重要低频系数按照位平面次序进行编码；最后根据零树结构对高频小波系数，从低分辨率开始逐频带搜索得到不为零的重要系数，对搜索到的高频重要系数同样按照位平面次序进行编码，直至最高频带小波系数。该算法的特点是具有可伸缩性，即其编码位流不但可以进行多分辨率解码，而且可以进行多位率解码。实验结果表明，该压缩方法性能优于Ｓｈａｐｉｒｏ的ＥＺＷ方法，能在一定程度上提高峰值信噪比（ＰＳＮＲ）。
In this paper, based on the characteristics of wavelet transform multiresolution decomposition and the advantages of EZW coding proposed by Shapiro, a new scalable zerotree coding method based on zerotree structure is proposed. The coding feature is that it can output multi-bitrate compressed bitstream with resolution, effectively Improved the EZW method can only be applied to a single resolution video service deficiencies. Firstly, the algorithm performs uniform scalar quantization on the coefficients of wavelet transform. Secondly, the important low-frequency coefficients which are not zero after quantization are coded according to bit-plane order. Finally, according to the zerotree structure, the high-frequency wavelet coefficients are searched by frequency band from low resolution Non-zero important coefficient, the search of the high frequency important coefficient in the same bit-plane coding order, until the highest band wavelet coefficients. The algorithm is characterized by scalability, that is, its encoded bit stream can not only multi-resolution decoding, but also for multi-bit decoding. Experimental results show that the proposed compression method is better than Shapiro’s EZW method and can improve PSNR to a certain extent.