Certain investigations on the effective VLSI implementation of the lifting based architecture of the discrete wavelet transform for image compression

Abstract

Discrete Wavelet Transform (DWT) has become the standard tool in most of the signal and image processing applications. DWT can provide notable compression ratios with a less loss of visual quality of the image, in comparison with those of the previous techniques. Further, image compression using DWT technique, avoids such limitations as obvious blocking artefact and poor subjective quality while the images are restored with higher compression ratios. The DWT has a pivotal role in decomposing the signal into different sub-bands with required time and frequency information. DWT has features, like region of interest coding, the ease of compressed image manipulation, progressive image transformation and so on. The lifting scheme of DWT is based on the spatial construction of the second generation wavelet that is most versatile in its factorization, and reduces the computations up to 50%, compared to convolution method-based DWT, which directly affects the memory, the power consumption, and the surface of the system newline