Extended performance comparison of tiling based image compression using wavelet transforms & hybrid wavelet transforms

The storage of images is becoming difficult with number of images growing to millions & billions. Hence the image compression is becoming absolute necessity in computing field. Use of the energy compaction property of orthogonal transforms is exploited for lossy image compression. In the recent research attempts better quality of compression is observed with the use of wavelet transforms & hybrid wavelet transforms. Further tiling based image compression is proven to be better with few hybrid wavelet transforms. The paper presents extended performance comparison of more variations of hybrid wavelet transforms for tiling based image compression. The experimentation is done on set of ten color images with size 256X256 each. Mean Square Error (MSE) between the original color image & compressed color image is employed to compare performance of various hybrid wavelet transforms for ten different assorted compression ratios (10%, 20%....90% & 95%). The hybrid wavelet transform generated using Cosine & Kekre Transforms (Cosine + Kekre) has shown the best performance with minimum MSEs for all compression ratios closely followed by the Hybrid Wavelet Transform generated using Cosine & Haar Transforms.