WWW Image Compression Resources

SPIHT IMAGE COMPRESSION

Some WWW Image Compression Resources

If you have a contribution that can be added to this page please send an e-mail to one address listed in the home page.

Compression FAQ

For introductory information about compression in general, a good place to start is comp.compression Frequently Asked Questions. (Note: some of its contents about the best compression methods are obsolete.) Links directly related to image compression are:

For information about medical imaging resources, try this FAQ.

Papers Related to SPIHT

An extensive study of compressed mammography image quality using SPIHT compression was performed at Stanford University. The paper can be downloaded in PostScript or portable document format (pdf) format.
SPIHT can also be efficiently used with DCTs, as shown by Z. Xiong, O. Guleryuz and M. T. Orchard, in their paper "A DCT-based embedded image coder."
Have you ever considered compressing an image defined on the surface of a sphere? Or any type of data in any manifold? Well, it can be done very efficiently with the proper multiresolution representation and SPIHT. See the papers about High-Dimensional Data Compression by Krasimir D. Kolarov and William C. Lynch
A new application for image compression proposed by Wm. Randolph Franklin. is described in Compressing Elevation Data (HTML version)
A method to exploit embedded-coding for error protection on noisy channels is proposed by P. Greg Sherwood and Kenneth Zeger.
The importance of trees in the EZW algorithm was studied by V. Ralph Algazi and Robert R. Estes Jr.. You can read a HTML summary or download a postscript paper.

Comparative Evaluations

The Waterloo BragZone, maintained by John Kominek, presents comparisons of several image compression schemes in various tests. Both numerical results and images are available.
See the compilation of PSNR results for the best Wavelet image compression methods, maintained by the Image Communications Lab at UCLA. (Among those listed, guess how many also have the same set of advantages.)
ARCHIVE1.COM is a site where you can compress with SPIHT interactively. The site has a large database of color images or you can upload your own images. You need a broadband connection to use this site effectively.
VcDemo is an Image and Video Compression Learning Tool developed by the Information and Communication Theory Group at Delft University of Technology that runs several image and video coders and displays the results before your eyes. With its intuitive graphical interface (Windows 95/98/NT), you can easily compare SPIHT against a host of other coders (EZW, VQ, fractal, JPEG,...) using your favorite images.

Wavelet References

Plenty of Wavelet Links.
MathSoft created a nice list of Wavelet resources

Compression Links

This set of compression pointers is an excellent place to start any search.
Links for Signal Processing and Compression resources, and an interesting paper too.

Some Other Interesting Papers

And when compression is not for "your eyes only"? See The Effect of Lossy Image Compression on Image Classification by J. D. Paola and R. A. Schowengerdt.
Want another different application of wavelet-based compression? Visit The FBI Fingerprint Image Compression Standard.
A slide presentation on the advantages of wavelet and SPIHT compression. Wavelet-Based Image Compression by Amir Said.

Programs

Need a simple program for lossless and also high-quality lossy compression? Try the Huffman-based program we use everyday, called lsp.

Is it good enough? Well, we used it for lossless compression in Stuart Inglis' se t of 9 test images ( Lena, Goldhill, Babboon, Camera, etc.) and obtained an average 1.72 compression ratio, while the state-of-the-art CALIC yielded 1.78 (only 3% larger!). And lsp it can be used for lossy compression too!
The JPEG images shown in our comparisons were obtained with the program XV by John Bradley. (Alternative link. )

Links to other pages.

SPIHT Home Page;

Properties of the Method

Licensing Information