An Image Compression Method Based on Subspace and Downsampling

Abstract

This study proposes a new Karhunen-Loeve transform-based algorithm with acceptable computational complexity for lossy image compression. The proposed study includes a simple algorithm using downsampling and KLT. This algorithm is based on an autocorrelation matrix found by clustering the highly correlated image rows obtained by applying downsampling to an image. The KLT is applied to the blocks of the downsampled image using the eigenvector matrix of the autocorrelation matrix, and the transform coefficient matrix is obtained. One of the important features of the proposed method (PM) is sufficient for a test image to have one transform matrix with low dimensional. The PM was compared with JPEG, BPG, and JPEG2000 compression methods for the Peak signal-to-noise ratio- Human