基于整数提升小波变换的改进EZW编码算法

IEZW Encoding Algorithm Based on Integer Lifting Wavelet Transform

为了提高图像有损压缩的性能,在分析EZW算法的基础上,提出了一种改进的零树小波编码算法。该算法采用整数提升小波变换对图像进行分解,将分解后的图像作改进的EZW编码,即将低频子带和高频子带分开进行编码,主表采用不同的扫描方式,同时取消副表中的极间扫描。试验结果表明,改进算法提高了图像压缩质量和峰值信噪比,同时减少了编解码时间。

Based on the analysis of the EZW (Embedded Zerotree Wavelet) algorithm, an improved zerotree wavelet encoding algorithm is proposed for improving the performance of lossy image compression.Integer lifting wavelet transform is used to carry out the decomposition of the image, and then the decomposed image is encoded by the improved EZW algorithm, in which the low and high frequency sub-bands are encoded separately and a different scanning method is used in the primary table while the inter-polar scanning in the assistant table is cancelled.Experiment results show that the proposed algorithm improves the Peak Signal Noise Ratio (PSNR) by more than 10.4 percent and reduces encoding and decoding time by more than 33.8 percent, compared to the original EZW algorithm.