摘要
提出了一种新颖的基于小波系数上下文模型的图像压缩方法,该方法通过量化当前系数的线性预测值形成上下文,进行自适应的算术编码;同时利用了小波变换的多分辨率性质,以渐近分辨率的方式压缩图片,具有分辨率可扩展性。实验结果表明,该方法获得的无损压缩比高于SPIHT和用于JPEG2000的EBCOT,在各分辨率下的压缩比也高于EBCOT,压缩时间也比EBCOT要少。
A new image-coding scheme based on context model of wavelet coefficients is proposed. The arithmetic encoder encodes wavelet coefficients with the contexts formed by quantizing linear prediction values. In addition, by exploiting the multi-resolution property of wavelet, the scheme that is resolution-scaleable can compress the transformed image for progressive resolution. Experimental results show that the scheme achieves higher lossless compression rate of image than lossless SPIHT and lossless EBCOT that is used in JPEG2000. The scheme earns higher compression rate for each scale of the image than EBCOT and costs less time to compress image than EBCOT.
出处
《微电子学与计算机》
CSCD
北大核心
2006年第4期153-155,159,共4页
Microelectronics & Computer
基金
重庆市科技攻关项目(2002CC23)
关键词
小波
上下文
无损图像压缩
渐进分辨率
Wavelet, Context, Lossless image compression, Progressive resolution
