光学学报, 2009, 29 (s2): 229, 网络出版: 2010-01-27  

基于优化整数小波变换的静止图像编码算法

Still Image Coding Algorithm Based on Optimal Integer Wavelet Transform
作者单位
北京师范大学 信息科学与技术学院, 北京 100875
摘要
内插双正交整数小波变换(IWT)支持高效的图像无损压缩并且具有较低计算复杂度,但是为了保证整数输出,变换中包含了浮点数缩放因子并额外增加了三个提升步骤,降低了整数小波变换对图像的有损压缩效率。提出了一种基于优化因子的静止图像编码算法。在小波变换过程中,新算法利用一组基于2的整数次幂的分数代替浮点数缩放因子,消除变换中的浮点数乘法操作,降低变换的计算复杂度。实验结果表明,采用优化因子的图像压缩算法不仅有效降低了编码中小波变换的计算复杂度,而且获得了与采用浮点数缩放因子的内插双正交整数小波变换相近的峰值信噪比。
Abstract
Interpolating biorthogonal integer wavelet transform (IB-IWT) can support efficiently lossless image compression and has low computational complexity. However, to ensure the integer output, the scaling factors of floating-point number and three additional lift steps must be included on the integer wavelet transform, which can reduce the lossy compression efficiency of IB-IWT. A new still image coding algorithm is presented based on optimal factors (OF). When the image is transformed, the new algorithm uses a group of fraction based on the integer powers of two instead of the scaling factor of floating-point number, which makes the multiple operations of floating number not to be included in the IB-IWT and reduces the computational complexity of IB-IWT. Experimental results show that the image compression algorithm based on OF not only reduces the computational complexity of IB-IWT in the image coding, but also provides the peak signal-to-noise ratio (PSNR) performance to be comparable to the IB-IWT using the scaling factors of floating number.

张立保, 余先川, 王鹏飞. 基于优化整数小波变换的静止图像编码算法[J]. 光学学报, 2009, 29(s2): 229. Zhang Libao, Yu Xianchuan, Wang Pengfei. Still Image Coding Algorithm Based on Optimal Integer Wavelet Transform[J]. Acta Optica Sinica, 2009, 29(s2): 229.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!