Varying proportions of constituent transforms to generate hybrid wavelet transform for image compression

The storage and transfer of images have gained importance. Image compression plays a vital role for ease of storage and transfer of images. Image compression in transform domain is one of the most popular techniques. Previous work has proved that the hybrid wavelet transforms (HWT) are better than the respective orthogonal transforms [1]. In this paper, the effect of varying proportions of constituent transforms to generate the hybrid wavelet transform are tested for the effect on quality of image compression using cosine and Kekre transform. The experimentation is done on set of 20 images by varying the proportion of constituent transforms and compression ratios (CR). Here, in addition to conventional 1:1 proportion of constituent transforms; four additional proportions alias 1:16, 1:4, 4:1, 16:1 are also considered for generation of HWT. The results prove that for compression ratio above 90%, 4:1 ratio of constituent DCT-Kekre transform in HWT gives better results. 1:1 ratio of constituent transforms give better results for compression ratios between 75% and 90%. But for lower compression ratios, 1:4 proportions of constituent transforms (DCT-Kekre transform) in HWT gives better results.