一种基于对象的分形视频压缩算法研究

基于分形视频压缩的典型方法,提出了一种改进并提高其压缩性能的方法,用更有效的宏块划分规则替代传统的四叉树划分规则、简化块的搜索策略和范围、使用类似H.264标准的I帧和减少重复计算等。提出了一种在分形视频压缩算法中的基于对象(OB)压缩的实现方法,可以对任意对象单独进行压缩编码,极大地提高压缩比和压缩速度,降低比特率并节省数据传输时的带宽。实验结果表明,相比典型的方法,本文提出的分形视频压缩算法,压缩比提高了近4倍,压缩速度快近10倍,图像质量提高了3～5dB,显著地提高了分形视频压缩算法的性能;同时,OB的压缩方法简单有效,压缩比和压缩速度又有大幅度提高,使分形视频压缩的应用具有更大的灵活性和实用性。

A research on object-based system for fractal video compression

Based on the classical fractal video compression method,an improved method is proposed,which uses more effective macroblock partition scheme instead of classical quadtree partition scheme;reduces the block searching strategy and range,thus increases the calculation speed greatly;uses homo-I-frame like in H.264;and reduces the repeating calculation,etc.And a new method of object-based fractal video coding is proposed in this paper,each object can be encoded and decoded independently,which results in increasi...