A High Speed VLSI Architecture for Lifting Based Discrete Wavelet Transform

Abstract: The Discrete Wavelet Transform (DWT) is well known for its application in image and video compression. Due to its remarkable advantage over the discrete cosine transform (DCT) in image compression, 2-D DWT has been accepted for the JPEG-2000 compression standard. The implementation of 2-D DWT, however, is highly computation-intensive. The lifting scheme is an efficient implementation of wavelet transform. Compared with convolution based DWT, lifting-based architectures not only have lower computation complexity but also require less memory. Modifications are made to lifting scheme in order to reduce critical path delay and size of temporal buffer. Keywords: Lifting scheme, Discrete Wavelet Transform, Discrete Cosine Transform, VLSI architecture