曲波变换和最小二乘支持向量机的图像压缩算法  

Image compression algorithm based on curvelet transform and least squares support vector machine

作　　者：刘康明[1] 

机构地区：[1]黄淮学院信息工程学院,河南驻马店463000

出　　处：《激光杂志》2014年第12期36-39,44,共5页Laser Journal

基　　金：国家自然科学基金资助项目(编号:61372058);辽宁省高等学校优秀人才支持计划项目(编号:LR2013012)

摘　　要：为了提高图像压缩质量,针对传统压缩算法的不足,提出一种曲波变换和最小二乘支持向量机相融合的图像压缩算法。首先采用曲波变换把图像分解为不同尺度和不同方向的曲波系数,并采用熵编码对粗尺度层曲波系数进行压缩,然后利用最小二乘支持向量机对细尺度层中不同方向的曲波系数进行学习,并通过和声搜索算法优化最小二乘支持向量机,实现细尺度层曲波数的压缩,最后采用图像压缩仿真实验测试其性能。结果表明,曲波变换和最小二乘支持向量机相融合的图像压缩算法提高了图像压缩的峰值信噪比,加快了图像压缩的速度,获得了更好的图像压缩效果。In order to obtain good compression results of image, a novel image compression algorithm combining curvelet transform and least squares support vector machine was proposed in this paper. Firstly, images were decom_posed by curvelet transform to get curvelet coefficients of different scales and orientations, and curvelet coefficients of the coarse scale were compressed by the entropy coding, and then the curvelet coefficients of different detail scales and orientations were learnt by least squares support vector machine and the parameters of least squares support vectors are optimized by harmony search algorithm, finally the performance is test by simulation experiments. The results indicate that the proposed algorithm can images the PSNR and fasten the compression speed, so it can obtain good compression efficient.

关 键 词：图像压缩 最小二乘支持向量机 曲波变换 和声搜索算法 

分 类 号：TN911.7[电子电信—通信与信息系统]