Slashdot Mirror
JPEG2000 Coming Soon
Sonny writes "In a few months time, internet users will be able to make use of the JPEG2000 standard which, its developers claim, enables web graphics to be downloaded much faster than is currently possible. This will not only make graphics-heavy web pages easier to download, it will also preserve image quality. The JPEG standard compresses image files which are then transmitted across the web faster than uncompressed files. Now, researchers at universities around the world have developed JPEG2000, the next-generation image-compression technology under the auspices of the International Standards Organisation. It is the first major upgrade of the standard since it first appeared in the early '90s. What is also important about the technology is its ability to send files without loss of data, which is not the case with current JPEG files. To take advantage of a JPEG2000, web browsers will need a Plug-In for either Internet Explorer or Netscape browsers. These free plug-in's are expected to be available later this year. The extension for the new files will be ".jp2"."
"If we aren't all using PNG right now, there's no way we're gonna be using jp2 "
.PNG wasn't adopted in the internet world is that it didn't compress enough. Also, it's alpha channel was never really utilized. There are those in the 3D-Art world that think .PNG is a god-send, however.
You're talking about the difference between 300k and 20k. The reason that
JPEG2000 has a few things going for it:
- Familiar Name
- Familiar Standard
- Smaller filesizes
- Likely to be better supported by IE and other browsers
"Derp de derp."
See this bugzilla entry for Mozilla's jpeg2000 progress.
Doesn't seem too promising:
If you look at appendix L of the jpeg2000 draft, there are 22 companies who believe that implementing the spec may require use of their patents.
PNG still hasn't taken off despite being supported in all major browsers (now if only IE did proper alpha, any year now...), how much chance does an image format that requires third party plugins have?
According to this pdf,
.9
the report compares 4 compression codecs, and found for a small sample found:
MEAN LOSSLESS COMPRESSION RATIOS (big is good)
------------------
JPEG 2000: 2.5
JPEG-LS: 2.98
L-JPEG: 2.09
PNG: 3.52
JPEG-LS is was usually the best, but PNG had a few really good sample that pushed its average up. Actually, these outliers appear important, because that is what really separates the codecs on this metric.
Lossless Decoding Times, relative to JPEG-LS (big is bad)
-----------------
JPEG 2000: 4.3
JPEG-LS: 1
L-JPEG:
PNG: 1.2
This doesn't make JPG2K appear too impressive. What it does offer, however, is features. Like Region Of Interest (ROI) coding, good lossy compression, random access, and other goodies that some people may really care about. The report claims that png doesn't do lossy encoding, which is news to me, but it does appear to be one of their major selling points for jpeg-2000 over png.
Hey, I've implemented a JPEG-2000 codec using
a BSD-style license.
It's been tested at the MIT biodmedical department already for compression of medical images.
It's available at http://j2000.org/.
It would be nice to see this work in my favourite browsers.
Several things, besides simply "good compression."
JP2 uses wavelet compression such that an image is effectively compressed at various resolutions below the originally, independently. Not only does this allow a high level of redundancy removal (which is why wavelets are good in the first place) and thus high compression, but jp2 tags each of these sections (subbands) separately in the compressed file.
So what? Well, a file with all of these sections is effectively a losslessly compressed image. However, this file can be further compressed (loss-ily) by simply throwing out some of these tagged sections! That is, you can make a "lossless" thumbnail image by keeping all the lower resolution subbands. Or, you can get a lower-quality (but smaller) fullsize version by throwing out some subbands at each resolution.
Better still, this manipulation can be done without decompressing the original image. Simply using only certain tagged sections of the file.
Consider this possible application of all this: Digital Cameras. A camera could take images at full resolution and lossless quality until the memory card starts filling up. Then, gradually as more and more room is required, it could quickly reduce the size or quality of previous pictures to make room for new pictures. Thus, you always have "enough" room for more pictures, provided you don't mind the quality reduction.
Of course, there are numerous uses for web applications -- thumbnails and full-sized images could be the same file, provided the web server knows how to parse the image file. (Little or no computation necessary, just sending parts of the file)
Anyways, JPEG2000 is very very cool.
But web pages aren't the only things digital images are used in. Think cameras.
This site illustrates the difference in quality between JPEG and JPEG2K. You get essentially a 5x reduction in storage space without losing quality, and the type of artifacts aren't as annoying, either.
Yes, the reference is on MSDN's site.
It's not complicated to use, it's just awkward, and you need to use PHP (or Javascript, or some other solution) if you want it to work in both IE in Mozilla. Here's an example of how I've done it in the past:
<img src="blank.gif" style="filter:progid:DXImageTransform.Microsoft.A
<? } else { ?>
<img src="imagewithalpha.png">
<? } ?>
The implementation of this in IE seems to be pretty good -- at least I haven't run into any problems with it.
Jason.