Wavelet Transforms

Serge Kazantzev The foundations of wavelet theory is today securely in place and a lot of work has been done over the last decade through a mix of theory and practice to make wavelets evolve with new applications. Wavelets are not only known for image compression (JPEG2000), but also for pattern recognition, process modelling and movies to name a few.

It looks natural to bring wavelets to Tcl. [While CL has enormous affection for both wavelets and Tcl, he's concerned that Tcl's arithmetic, relatively distant from the hardware, excessively handicaps performance. In particular, it's clumsy to do all the power-of-two calculations that make wavelets so apt for computerization. On the other hand, it's been a decade since CL has thought much about wavelets, so maybe these difficulties have all been surmounted.]

YawTcl - a package that uses C/C++ code to do the math - is a modest attempt to give Tcl a subset of the Cohen-Daubechies-Fauveau family of the biorthogonal wavelets to play with, and this wiki page is set up to share experience and practice.

CL's concern about performance is certainly a valid point. The use of TEA or critcl is a way to address the performance issue: give the calculations to a 'low-level' language such as C/C++, and let Tcl play with a 'meta' object.