无线电工程

为了解决主成分分析算法在图像压缩中不能有效提取非线性特征问题,采用核主成分分析算法对原输入空间进行非线性变换将其映射到特征空间,在特征空间下使用主成分分析算法提取特征从而达到了对非线性特征的提取。通过累计贡献率选取主成分实现数据压缩的目的。采用峰值信噪比和压缩比分别评定图像压缩质量和压缩程度。仿真结果表明,在相同的累计贡献率下核主成分分析的峰值信噪比和压缩比分别提高了约2倍和1.6倍,并有更好的非线性特征提取能力。通过与传统的JPEG对比,该方法具有更好的性能,从而说明核主成分分析算法适用于图像压缩。

In order to solve the problem that the principal component analysis algorithm can't extract the nonlinear feature effectively in the image compression,the kernel principal component analysis algorithm is used to map the original input space to the feature space.The feature space is extracted by the principal component analysis algorithm in the feature space,thus the nonlinear feature is extracted.The principal component is selected by accumulative contribution rate to achieve the purpose of data compression.The peak signal to noise ratio and compression ratio are used to evaluate image compression quality and compression degree respectively.The simulation results show that the peak signal to noise ratio( SNR) and the compression ratio of the kernel principal component analysis( PCA) are increased by about 2 times and 1.6 times respectively at the same cumulative contribution rate,and have better nonlinear feature extraction capability.Finally,it has better performance compared with the traditional JPEG,which shows that the PCA algorithm is suitable for image compression.