基于Y-H模型的奇异值分解图像压缩方法

摘要

It is required to preserve the relativity information of each channel for color image when it is compressed. This paper proposes acompression method based on a new style of Singular Value Decomposition(SVD) founded on Young-Helmholtz model--YH-SVD. It proofs theexistence and gives out the structural style of YH-SVD, and applies this method to compress color image after it is transformed into Greaves space. Experimental results show that this method can keep the clearness of reconstructed image, and can achieve high Peak Signal to Noise Ratio(PSNR).%彩色图像在压缩时需要保存通道间的关联信息.针对该情况,提出一种基于Young-Helmholtz模型的奇异值分解(YH-SVD)图像压缩方法.证明YH-SVD的存在性,求得其结构形式,使用Greaves对Young-Helmholtz模型进行变换,由此进行图像压缩.实验结果表明,利用该方法能获得清晰的重构图像,峰值信噪比较高.