Hi Matthias, I have created a bug for your issue: http://bugs.mitk.org/show_bug.cgi?id=8147 We will check it during one of the next bug squashing parties.
Best Regards, Ingmar -----Ursprüngliche Nachricht----- Von: Matthias Keil [mailto:[email protected]] Gesendet: Dienstag, 10. Mai 2011 13:46 An: [email protected] Betreff: [mitk-users] Erroneous transformation when writing .mhd files Hi there, I am currently facing a problem while writing segmentations as .mhd files. I am using a Subversion version of MITK (Revision 29842; the latest in the old Subversion Repository) together with ITK 3.20.0. When I am loading an .mhd dataset with the following header information: ObjectType = Image NDims = 3 BinaryData = True BinaryDataByteOrderMSB = False CompressedData = False TransformMatrix = 1 0 0 0 1 0 0 0 1 Offset = 31.395 21.252 16.65 CenterOfRotation = 0 0 0 AnatomicalOrientation = RAI ElementSpacing = 0.483 0.483 0.49 DimSize = 223 211 122 ElementType = MET_UCHAR ElementDataFile = MyFilename.raw And save it again as a .mhd file, I am getting the following header information: ObjectType = Image NDims = 3 BinaryData = True BinaryDataByteOrderMSB = False CompressedData = False TransformMatrix = 107.709 0 0 0 101.913 0 0 0 59.78 Offset = 31.395 21.252 16.65 CenterOfRotation = 0 0 0 AnatomicalOrientation = RAI ElementSpacing = 0.483 0.483 0.49 DimSize = 223 211 122 ElementType = MET_FLOAT ElementDataFile = MyFilename2.raw Please note the difference in the TransformMatrix line. For some reason the writing with my current MITK QT4 version adds some rotation which is not in the data. The old QT3 version did not do that. >From my testing, I can say that: - writing the .mhd file adds this arbitrary transformation - this transformation is not visible in the "details" view in the Datamanager (there, only the scaling values are visible) - this transformation is not being read or used by both the old QT3 and the new QT4 MITK version when loading the newly written dataset (both, the new and the original one are at the same location and orientation) I am guessing that this is error is either related to MITK (handing some transformation to the .mhd file writer which is not actually present) or to ITK who are now writing transformations to the .mhd files which they didn't in version 3.16.0 that I have used with MITK QT3 before. Anyway MITK should cause the file writer to write the identity matrix in my case. Could you please try to reproduce this error with your current MITK version? Thank you very much, Matthias -- Dipl.-Ing. Matthias Keil Research Assistant 'Cognitive Computing & Medical Imaging' Fraunhofer-Institut für Graphische Datenverarbeitung IGD Fraunhoferstr. 5 | 64283 Darmstadt | Germany Tel +49 6151 155-212 | Fax +49 6151 155-480 [email protected] | www.igd.fraunhofer.de ------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ mitk-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mitk-users ------------------------------------------------------------------------------ What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ mitk-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mitk-users
