摘要：The purpose of this paper is to introduce an image compression scheme using a combination of wavelet packet transform and vector quantization. All the wavelet packet bases corresponding to various tree structures have been considered and the best one has been coined based upon the peak signal to noise ratio and compression ratio of the reconstructed image. In first step input image decorrelation using the wavelet packet transform, the second step in the coder construction is the design of a vector quantizer. Vector Quantization (VQ) is fast and efficient method of quantizing Laplacian-like data, such as generated by transforms (especially wavelet transforms) or sub-band filters in an image compression system. VQ has very simple systematic encoding and decoding algorithms and does not require codebook storage. VQ has culminated in high performance and faster VQ image compression systems for both transforms and subband decompositions. The proposed algorithm provides a good compression performance The introduction of wavelets gave a different dimension to the compression. But there are some limitations of wavelets while handling the line and curve singularities in the image. There are transforms beyond wavelets namely – Curvelet and Ridgelet Transforms. This paper aims at the analysis of compression using Curvelet, Wavelet and the Ridgelet Transform. The Curvelet Transform gives better performance in terms of PSNR. Wavelet performs the least and is also affected by the blocking artifacts. By selecting proper thresholding method, better results for PSNR have been obtained. Using the proposed algorithm the compression ratio is increased by about 25 – 30 % as compared to conventional DWT technique.
