For Programmers: Free Programming Magazines  


Home > Archive > Compression > February 2005 > Re: Compression ratio.









You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

 

Author Re: Compression ratio.
Martin Brown

2005-02-12, 3:55 pm

Christian Gollwitzer wrote:
> Uni wrote:
>
> Hmm? A jpeg file (more accurately JFIF) contains the quantization tables
> used for the compression, because they are needed to decompress the
> image. These tables are the only exhausting source for the "compression
> quality" of the jpeg image. The construction of the qtables from the
> "quality value" shown to the user is arbitrary and depends on the
> compressor. You are right, that this "quality setting" is normally not
> saved, but even if it would be, it could not compared between applications.


Quite a few applications do use unimaginatively scaled versions of the
original "sample" table given in the appendix of the JPEG standard. So
with a bit of cunning you can compute a very good guess at the original
quality setting from the quantisation tables in a JPEG image.

It only really breaks down if the quantisation tables are customised.

PhotoShop is a notable exception using its own symmetrical custom
quantisation tables. I recently saw a Nikon scanner JPEG that had all
values in the quantisation table idenically equak to 1.

Regards,
Martin Brown
Sponsored Links







Also available: Server administration forum archive | Web Design forum archive | Software forum archive | Hardware reviews archive

Copyright 2008 codecomments.com