An Adaptive Lossy Image Compression using Singular Value Decomposition and Wavelet Difference Reduction

Abstract: Image generation, sharing and analysis play an important role for day-to-day life. An adaptive lossy image compression using singular value decomposition (SVD) and wavelet difference reduction (WDR) is proposed for the application of colour and gray scale images. The goal is to improve the quality and minimize the storage requirement. The proposed adaptive lossy image compression is composed of three main components. The first is the application of SVD to the original input image. Next the wavelet difference reduction method is applied to the original image. Both the compressed images are fused in the final step. Such fusion of two methods increases the quality of the reconstructed images. The performance of the proposed compression method is evaluated using the parameters namely compression Ratio (CR) and Peak-Signal-to-Noise Ratio. Experimental result shows that the proposed adaptive lossy image compression using fusion of SVD and WDR outperforms the cascading of SVD and WDR.

Keywords: Image Compression, Colour Images, Gray scale Images, Singular Value Decomposition (SVD), Wavelet Difference Reduction (WDR), Fusion.