Region-based wavelet transform for image compression

Wavelet transform is not applicable to the arbitrarily shaped region (or object) in images, due to the nature of its global decomposition. In order to solve this problem, the region-based wavelet transform (RWT) is proposed and its computational complexity is examined in this brief. It is shown that the RWT requires significantly fewer computations than conventional wavelet transform, since the RWT processes only the object region in the original image. Experimental results show that any arbitrarily shaped region in images can be decomposed using the RWT and perfectly reconstructed using the inverse RWT. Furthermore, the RWT outperforms the shape-adaptive wavelet transform (SAWT) in PSNR at the same compression ratio since the former generates less high-frequency information.