[OSGeo-Discuss] Open File Formats and Proprietary Algorithms

[Michael P. Gerlek]

Landon asked:

> When you said "there is today no open source implementation of JP2 that is suitable for geo
> work" do you mean that there is no open source library that can read and write JP2? If so,
> who is using the format?

There are a few implementations of JP2 around.  The Kakadu library, which is extremely compliant and featureful and robust (and correspondingly extremely big and complicated and scary) is the best-known package: it is available only via licensing fees.  LizardTech uses Kakadu, in fact, and a number of geo vendors use either Kakadu directly or LizardTech's packaging of it.

The ER Mapper folks had a JP2 solution at one time, but I never understood their licensing terms to be OSI compliant -- and since they got bought out by Leica I've sort of stopped tracking that issue.  If anyone has any current info, I'd like to hear it.

There are a couple truly open source libraries, but none have been written in such a way as to be able to support geo-sized imagery (>500MB, say).  Doing the wavelet algorithms efficiently for large data sets requires rocket science.


> Do you know why there hasn't been a broader adoption of JP2?

Not through lack of trying on my part :-)

I think the two biggest reasons are:

(1) The algorithms for handling large images in memory really are rocket science, and no one in the FOSS community has gotten the "itch" sufficiently bad enough to go and do the work needed inside the existing open source packages.  Hopefully someday someone will.

(2) MrSID (and, perhaps, ECW) are widely used and supported.  Philosophical motivations aside, MrSID and ECW have historically gotten the job done and so the need for JP2 just isn't as high as it otherwise might be.

That said, NGA is a good counter-example.  They support JP2 in a number of areas already and have mandates to broaden that support. 

-mpg