Publications

Abstract : Embedded Zero-tree Wavelet (EZW) is a wavelet based image compression scheme. It is basically a quantization stage that incorporates some characteristics of the wavelet decomposition. The EZW approach and its descendants significantly outperform some of the generic approaches. The wavelet coefficients in different sub bands represent the same spatial location in the image. This is an important characteristic used by EZW. In case of decomposition, since the size of the different sub bands is different, then a single coefficient in the smaller sub band may represent the same spatial location as multiple coefficients in the other sub bands. This paper describes hardware implementation of EZW encoding algorithm along with Huffman encoding and decoding architectures. After performing lifting based DWT technique and EZW algorithm, Huffman coding ensures further compression of the image. In Huffman coding no bit string is a prefix of any other bit string. Hence each code is uniquely decodable.