which TIFF's are supported by imread?

10 views (last 30 days)
Borislav
Borislav on 9 Jul 2012
On the Matlab help page about imread
there are a paragraph and a table with format-specific information about the TIFF format. The table is supposed to list the supported TIFF by bit/sample and compression type. The entries in the table are either y or -. At first I thought that y means "supported" and - means "not supported". However, I noticed that the 12-bit and 16-bit rows hold only -'s. That would mean that the 12-bit and 16-bit formats are not supported at all. But then why including those rows in the table if they are not completely unsupported. I would greatly appreciate if anybody can explain with certainty how that table is to be read. Thanks.
  3 Comments
Image Analyst
Image Analyst on 10 Jul 2012
Then what benefits, if any, does the Tiff class have over imread()? I guess if you need to extract out some tiff tags (meta data) for example? But if you don't need to do that, then just use imread()?
John
John on 10 Jul 2012
For most users, IMREAD is probably the better tool for reading TIFFs because it requires the fewest steps to get what most people would want. There are a couple of cases where the behavior of IMREAD differs from that of the Tiff class, such as in reading CIELab, YCbCr, MinIsWhite, or LogLuv images.
The big benefit to the Tiff class is in creating TIFF content. IMWRITE only gives you access to a very limited number of tags, whereas the TIFF class handles over 70, some of these being "pseudo tags" that control how various libtiff codecs work. The Tiff class also allows you to define tiled TIFFs and create subdirectories, where IMWRITE does not.

Sign in to comment.

Answers (2)

Image Analyst
Image Analyst on 9 Jul 2012
I don't know, but maybe you can try the newer "Tiff" class - it probably handles more flavors of tiff than the generic imread(), though I'm not sure.

Walter Roberson
Walter Roberson on 10 Jul 2012
Generally speaking, the tables list all formats supported by the standards for that image type. The rows that contain only unsupported entries are needed in order to show that they are unsupported, as otherwise people would be left not being sure.

Tags

Products

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!