Hybrid DWT-SVD-VQ Image Compression for Monochrome Images

This paper investigates the new method of combined image compression from the well known Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD). The method makes use of the energy compaction properties of both the transforms. On a sub block of an original image 2D wavelet transform is applied. On the transformed image the singular value decomposition is carried out. The resultant data contains maximum energy of the image in a minimum number of coefficients. These coefficients along with the singular matrix are Vector Quantized (VQ) and transmitted. Inverse operations are carried out in order to decompress the data. Thus this paper achieves high compression by exploiting the energy compaction properties of the best transforms that are applicable for image compression. Simulation results indicate that the compression obtained posses good subjective quality of image.