Small difference in time intervals

[albinahlback]

Hello,

I am currently writing software to interpret raw CZI-files for your microscopes and I am having trouble understanding some parameters within the files.

In a CZI-file, there is a metadata section as well as an attachment section, where the latter has a timestamp segment which contains timestamps for when the frames are taken. Moreover, in my setup we have a segment in the metadata which describes the time increment for the frames. For us, this can be found under

<ImageDocument>
  <Metadata>
   <Information>
    <Image>
     <Dimensions>
      <T>
       <Positions>
        <Interval>
         <Increment>

but perhaps this is specific for my setup. Our camera is Axiocam 305 and software is ZEN (blue edition) version 3.3.89.0000.

The problem is that there is a small difference in the calculated time differences in the timestamps versus the time increment in the metadata. The relative difference is about 10^-5, which is small, but far larger than any error that is expected from when converting from binary to decimal and vice versa.

Which should I trust the most?

Also, is there any up-to-date CZI design specification? Mine is quite outdated as it is from 2011, and the current state of the online documention does not seem complete. If there is one, would it be possible to send one to me?

Best,
Albin

[ptahmose]

The best answer or recommendation I could give here is:

• I assume you are refering to an attachment of "Name=TimeStamps", "ContentGuid={62F8C9BD-86C2-472C-84C0-E4D03DBF6360}" and "filetype=CZTIMS". If such an attachment is present, then it should be given precedence over the information in XML-metadata.
• However - be aware that there are no guarantees - e.g. the timestamps may not be equidistant.
• Afaik, it is not explicitely stated in CZI which information is authorative, and the XML does not state that a timestamp-attachment is to be considered or something. So, you'd need to check for the presence of the timestamp-attachment on your own.

[albinahlback]

Thanks, that answers my question! Is it possible to obtain a newer CZI specification?

[FelixS90]

Hi @albinahlback !

Please feel free to register at https://www.zeiss.com/microscopy/en/products/software/zeiss-zen/czi-image-file-format.html for free to get (a licensed) access to the CZI documentation. A more straightforward and intuitive (but in German) way is https://www.zeiss.de/mikroskopie/produkte/mikroskopsoftware/czi/czi-download.html.

Please let us know if that answers your question.

[albinahlback]

Sorry for the late reply. This does answer my question, but this is way too tedious to go through (I thought it was tedious enough to download the ZEN software). Is it really necessary to keep the CZI specification closed source?