[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Reply to: [list | sender only]
Simple images
- To: Multiple recipients of list <[email protected]>
- Subject: Simple images
- From: [email protected] (J.Campbell)
- Date: Mon, 27 Nov 1995 07:05:54 -0500 (EST)
Dear All,
I agree, in general, with Andy's answers to his own questions in his message
of 23/11/95. A binary file with ascii header information would seem to
be appropriate.
One question asked is 'What is simple?'. Here is my answer for PX rotation
or Laue images:
An cartesian image format with:
1) ASCII header information No. of pixels along fast axis.
No. of pixels along slow axis.
Raster dimensions.
Axis order with respect to some standard set of
laboratory axes.
The storage format of the binary image data.
2) The image binary data stored as two byte or four byte integer values.
3) One image/file.
Other thoughts in brief.
1) We seem to have managed reasonably well for quite a long time without
a standard format which suggests to me that the basic requirements
may really be quite simple.
2) A lot of other header information is possible but how will it be gathered
to put into the file as it may well not be available to the system
writing the image files.
3) Some form of 'squashed' two byte integer data would problably be very
useful (e.g. as used in the MOSFLM package).
4) Perhaps overload/bad pixels need to be identified in some specific way.
5) As the complexity of any format definition increases, e.g. data
compression, many different storage formats etc., it will be
increasingly important to provide the software to read such images
and not just make a format specification.
John Campbell
Reply to: [list | sender only]
- Prev by Date: imgCIF Bulletin board
- Next by Date: some comments
- Prev by thread: some comments
- Next by thread: imgCIF Bulletin board
- Index(es):