J4

• 研究论文 • 上一篇    

基于分形的快速压缩编码方法

齐利敏;刘文耀;吕大伟
  

  1. (天津大学 精密仪器与光电子工程学院,天津 300072)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-04-20 发布日期:2008-03-28

Faster compress coding arithmetic based on fractal

QI Li-Min;LIU Wen-Yao;L Da-Wei
  

  1. (Tianjin Univ. College of Precision Instrument and Opto-electronics Eng., Tianjin 300072, China)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-04-20 Published:2008-03-28

摘要: 分形编码是一种很有潜力的编码方法,但是基本的自动分形算法计算量大,编码时间长.针对此缺点,根据分层编码的思想,给出了一种具体的分形与简单量化编码相结合的加速分形编码方法.首先对原始图像进行1/4减采样,得到一幅减采样图像,利用基本的分形编码方法编码该图像,然后对所得到的编码在原始图像分辨率下进行解码,对解码图像与原始图像求差值,用简单的量化编码方法编码差值图像.差值编码与用基本方法所得到的压缩编码共同构成原图像的编码.与传统的分形方法相比,该方法不仅提高了编码时间,并且在信噪比、压缩比等方面得到了不同的改善.

关键词: 图像压缩, 分形, 迭代函数系统, 分辨率

Abstract: Fractal image compression is an efficient technique for encoding images. It can lead to a high compress ratio and decoding quality in theory. But, the speed of basic automatic fractal encoding is slow and the decoding quality is also unsatisfactory when the compress ratio is higher.This paper gives a way to get a fast speed of encoding. We reduce the original image to the 1/4 resolution, encode the 1/4 resolution image by the basic fractal algorithm, get the fractal code of the 1/4 resolution image, and decode the code to the original resolution. Then we calculate the difference between the decode image and the original image, and encode the difference image in a simple way. The code of the 1/4 resolution and the code of the difference image constitute the code of the original image.This method can speed up about 3-4 times with the same PSNR as for the basic automatic fractal algorithm.

Key words: image compression, fractals, iterated function system, resolution

中图分类号: 

  • TN919.8