iannellog
commented
6 years ago 2018-04-26 13:17 GMT+02:00 Alessandro Crimi notifications@github.com:
First, I'd like to compliment you on your useful software.
I am having trouble as the software is properly stitching 2 series of tif images. However, for some strange reason the first 60 slices are spread every 11 slices in the stitched version. The initial images are properly distributed so the bug(if this is a bug) is happening during stitching.
sorry, but I do not understand which is exactly your problem. What does it mean "the first 60 slices are spread every 11 slices in the stitched version"? Perhaps it is a problem of file names (see also comments below). Apart trying to explain more, could you please send me all the xml files generated by TeraStitcher and also the mdata.bin generated in the root directory of the dataset? Another question is: have you organized the dataset in the two-level hierarchy of directories and then imported in TeraStitcher or have you externally generated an xml import file? From your mail I understand that you have followed the first alternative. Is it right?
The data are the following:
2 series of tif images. Each file is 2048x2048 single channel. Each series is 352 images. voxel size: 6.5x6.5x10 um I am using as origin (40.836,0 I have no real reason but I have seen it in another example), which is leading me to two folders 408360 and 508360, resulting in a tile overlap 509(X), 0(Y)
if you import with TeraStitcher directly from the two-level hierarchy of directories and the nominal overlap between the two tiles are 509 pixels along X (that I understand is the horizontal dimension that we referto as "H") you should:
- have a two level hierarchy of directories with:
- one directory named 000000 at the first level
- two directories named 000000_000000 and 000000_100035 at the second level (i.e. inside 000000)
- a series of 352 images in each directory at the second level; the files should be named in such a way that they are lexicographically ordered according to their relative position (e.g. 000.tif, 001.tif, 002.tif, ... , 350.tif, 351.tif)
This way your origin will be 0,0,0.
I show one of the resulting stitched images
Is this image what you expect or it is somehow wrong? In the second case which is the problem? Another perhaps useful information is the final size of the stitched image (TeraStitcher should have created a folder named "RES(VVVVxHHHHxDDDD", which values are VVVV, HHHH, and DDDD?)
Let me know if this can help or send me more information.
Best.
-- Giulio
Giulio Iannello Preside della Facolta' Dipartimentale di Ingegneria Universita' Campus Bio-Medico di Roma v. Alvaro del Portillo, 21 00128 Roma, Italy
Tel: +39-06-22541-9602 E-mail: g.iannello@unicampus.it Fax: +39-06-22541-9609 URL: https://scholar.google.it/citations?user=L-UJxIgAAAAJ
alecrimi
First, I'd like to compliment you on your useful software.
I am having trouble as the software is properly stitching 2 series of tif images. However, for some strange reason the first 60 slices are spread every 11 slices in the stitched version. The initial images are properly distributed so the bug(if this is a bug) is happening during stitching. The data are the following: 2 series of tif images. Each file is 2048x2048 single channel. Each series is 352 images. voxel size: 6.5x6.5x10 um I am using as origin (40.836,0 I have no real reason but I have seen it in another example), which is leading me to two folders 408360 and 508360, resulting in a tile overlap 509(X), 0(Y).
As search region in the alignment step I am using 25(x)x25(y)x10(z) and overlap 509(x) and 50(y).
Any idea what is wrong? or how to fix this eventual bug? I show one of the resulting stitched images