计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (14): 156-159.

• 图形图像处理 • 上一篇    下一篇

基于DWT-LSSVM的图像压缩算法

褚  静,徐安成,张美凤   

  1. 常州工学院 光电工程学院,江苏 常州 213002
  • 出版日期:2013-07-15 发布日期:2013-07-31

Image compression algorithm based on discrete wavelet transform and least square support vector machines

CHU Jing, XU Ancheng, ZHANG Meifeng   

  1. College of Optoelectronic Engineering, Changzhou Institute of Technology, Changzhou, Jiangsu 213002, China
  • Online:2013-07-15 Published:2013-07-31

摘要: 为了进一步提高图像压缩效率和质量,提出一种离散小波变换(DWT)和最小二乘支持向量机(LSSVM)相融合的图像压缩方法(DWT-LSSVM)。采用DWT对图像分解,得到低频系数和高频系数,采用LSSVM归学习逼近高频系数,并采用混沌粒子群算法对LSSVM参数进行优化,对支持向量、权重和低频系数进行编码,得到数据压缩数据流。仿真结果表明,DWT-LSSVM获得了较高的压缩比,可以较好满足图像传输的实时性要求。

关键词: 图像压缩, 离散小波变换, 最小二乘支持向量机, 嵌入式零小波算法

Abstract: In order to further improve the image compression efficiency and quality, this paper puts forward an image compre-
ssion method based on Discrete Wavelet Transform(DWT) and Least Squares Support Vector Machines(LSSVM). DWT is used to decompose the image to get the low-frequency coefficients and high frequency coefficients, and then LSSVM which parameters are optimized by chaotic particle swarm algorithm is used to train the high frequency coefficients, the support vector, weight and low frequency coefficients are encoded to form the data compression stream. The simulation results show that the proposed method obtains higher compression ratio and can satisfy the real-time requirement of image transmission.

Key words: image compression, discrete wavelet transform, least squares support vector machines, embedded zerotree wavelet algorithm