From Lossy to Lossless Wavelet Image Coding in a Tree-Based Encoder with Resolution Scalability
For a lossy encoder, it is important to be able to provide also lossless compression with little or no modification of the usual algorithm, so that an implementation of that algorithm can work in lossy or lossless mode, depending on the specific application, simply by varying the input parameters. In this paper, we evaluate the capability of the Lower Tree Wavelet (LTW) image encoder to work in lossless mode. LTW is a fast and multiresolution wavelet image encoder, which uses trees as a fast mode to group coefficients. In addition, general details on how to implement efficiently (i.e., with only shift and addition/subtraction operations) a reversible integer-to-integer wavelet transform are also given, as a requirement to implement a wavelet-based lossless encoder. Numerical results show that despite being general purpose (i.e., both lossy and lossless) and lacking of complex techniques (such as high-order context and predictive coding), the LTW performs as well as JPEG 2000 in lossless mode, and only 5% below LOCO-I, a specific lossless algorithm.
KeywordsWavelet Coefficient Image Encoder Predictive Code Lossless Compression Lift Scheme
- 3.ISO/IEC 15444-1: JPEG2000 image coding system (2000)Google Scholar
- 4.Said, A., Pearlman, A.: A new, fast, and efficient image codec based on set partitioning in hierarchical trees. IEEE Transactions on circuits and systems for video technology 6(3) (June 1996)Google Scholar
- 5.Shapiro, J.M.: Embedded image coding using zerotrees of wavelet coefficients. IEEE Transactions on Signal Processing 41(12) (December 1993)Google Scholar
- 6.Oliver, J., Malumbres, M.P.: Fast and Efficient Spatial Scalable Image Compression Using Wavelet Lower Trees. In: Proc. IEEE Data Compression Conference, Snowbird, UT (March 2003)Google Scholar
- 11.Antonini, M., Barlaud, M., Mathieu, P., Daubechies, I.: Imagen Coding Using Wavelet Transform. IEEE Transactions on Image Processing 1(2) (April 1992)Google Scholar