AWARE SYSTEMS
TIFF and LibTiff Mail List Archive

Thread

2004.12.15 04:30 "[Tiff] Q: Fax photometric", by Chris Losinger
2004.12.18 17:44 "Re: [Tiff] Q: Fax photometric", by Bob Friesenhahn
2004.12.18 18:40 "Re: [Tiff] Q: Fax photometric", by Joris
2004.12.18 20:58 "RE: [Tiff] Q: Fax photometric", by Bill Bither
2004.12.18 23:13 "Re: [Tiff] Q: Fax photometric", by Joris
2004.12.18 21:24 "Re: [Tiff] Q: Fax photometric", by Bob Friesenhahn
2004.12.20 13:34 "Re: [Tiff] configuring for JBIG", by Leonard Rosenthol

2004.12.18 17:44 "Re: [Tiff] Q: Fax photometric", by Bob Friesenhahn

On Tue, 14 Dec 2004, Chris Losinger wrote:

I have an image that's CCITTRLE compressed, with a photometric tag = 1 (black=0)

        the TIFF spec says:

The "normal" PhotometricInterpretation for bilevel CCITT compressed data is
WhiteIsZero. In this case, the CCITT "white" runs are to be interpretated
as white,
and the CCITT "black" runs are to be interpreted as black. However, if the
PhotometricInterpretation is BlackIsZero, the TIFF reader must reverse the
meaning
of white and black when displaying and printing the image.

so, according to the spec, this should be interpreted as black=0 -- photometric tag overrides CCITT default. but, if i do that, the image is clearly inverted (it's a fax, and the background should be white).

however, Photoshop 6 and 8, and MS Document Imaging all read it so that it looks right, white=0. LibTiff reads it 'inverted' (incorrect in appearance, but apparently correct according to the spec: black=0).

There is a choice between doing what the specification says should be done and the most common usage. It seems that these popular applications always assume the normal case. They may do this because some TIFF files are created with the wrong setting and users complain.

I believe that current GraphicsMagick behaves similarly to libtiff.

Bob
======================================
Bob Friesenhahn
bfriesen@simple.dallas.tx.us
http://www.simplesystems.org/users/bfriesen