Posted by – Wednesday 2012-05-02
In the blog post “Adding ECW support to GDAL via plugins” (February 2011) I shared the experiences I had to get the GDAL library able to work with this proprietary format.
Some days ago I had to repeat the task, and that recipe did not work. Under Debian 6 I was unable to build libecwj2-3.3.tar.gz for GDAL 1.6.3 using GCC 4.4.5. One year ago the environment was different: Debian 5, GDAL 1.5.2 and GCC 4.3.2.
Fortunately, via Google I found “Patch to build libecwj2-3.3.tar.gz with GCC 4.1″ on OSGeo’s gdal-dev mailing list, which gave me a clue to solve the problem.
Posted by – Sunday 2011-02-27
In this post we will see how to add to the Geospatial Data Abstraction Library  (GDAL) library limited support [note 1] for ER Mapper’s Enhanced Compression Wavelet (ECW) format. The environment will be an Ubuntu Linux operating system.
The ECW format has become very popular because it achieves high compression ratios at the expense of small amounts of RAM memory and computing time. Another remarkable feature of this format is selective decompression, that is, it is “able to selectively decompress a portion of digital imagery while viewing the image, and to selectively decompress the image at different levels of detail (as the user zooms in or out the image view)” . Other popular compression formats like TIFF or JPEG do not feature selective decompression; the whole image must be decompressed before viewing it, a seriuos issue when working with gigabyte sized images like orthophotographs.
The ECW format is propietary and, thus Linux distros like Debian or Ubuntu do not provide support for ECW in GDAL, even in their selections of non-free packages. That is the reason of the ‘install recipes’ shown in this post.
In the first section we will see how to compile this library, creating a .deb package useful for future installations. In the second section we will see how to install the GDAL’s ECW plugin, also needed to get GDAL able to read ECW files.
“Come abilitare il supporto ai file ECW in GDAL e visualizzarli in GRASS e QGIS“, written in Italian language, is the article this one is based on.