Computing That Serves

Length-Limited Data Transformation and Compression


Thursday, February 10, 2005 - 10:00am


Joshua Senecal, Doctoral Candidate, University of California

We are investigating fast and efficient data transformation and compression methods where data values are not permitted to exceed a certain number of bits. This makes the process more compact, and allows it to take place in environments, such as graphics hardware, where resources are limited. We developed a prototype data compressor using length-limited variable-to-variable length codes, and results with the prototype indicate that its speed and compression rate are competitive with state-of-the-art coders. We also developed the Piecewise-Linear Haar (PLHaar) transform, which is to our knowledge the only n-bit to n-bit reversible transform suitable for lossy and lossless coding. PLHaar is conceptually simple, easy to implement, and delivers lossy image reconstructions that are more visually appealing than those produced by similar transforms.


Joshua Senecal received the BS degree in Computer Science from Revelle College at the University of California, San Diego, in 1999. He is a PhD candidate at the Institute for Data Analysis and Visualization at the University of California, Davis, and is currently a research fellow at the Institute for Scientific Computing Research, Lawrence Livermore National Laboratory. His interests include data compression, image coding, image processing, scientific visualization, and computer graphics. His current research focuses on developing fixed-width image transforms and data compression techniques to support scientific visualization and high-performance computing. He is a student member of the IEEE and the ACM.