Content-type: text/html Manpage of CT-IPC


Section: User Commands (1)
Index Return to Main Contents


ct-ipc - Forms IPC-based sinogram(s).  


ct-ipc [OPTIONS...] <far-intensity> <0-intensity>  


Performs following procedures:
1) reads text file(s) describing the array(s) of input files (foreground-background pairs) in contact print plane and at a distance from the object.
2) removes the backgrounds in accordance with these files.
3) performs the IPC extraction of the requested contrast component.
4) constructs the sinograms of the slices requested in the slice string.



List of input images taken at the distance.
List of input images taken in the contact print plane (clean absorption contrast).
Arguments "<far-intensity>" and "<0-intensity>". Text file which describes the stack of input images: foregrounds and (optionally) background(s). The strings in the file can be one of the following forms:

    #BACKGROUND# <background filename>
The string prefixed with the "#BACKGROUND# " gives the name of the background file right after the prefix. This background file contains the background contrast for all foregrounds below it until the next background string is met. No background removal is performed for all foregrounds given in the strings above the first background. If the string starting with the prefix does not contain any background name then all following foregrounds are considered as is ( i.e. no background removal is performed).

    #DARKCURRENT# <dark current filename>
The string prefixed with the "#DARKCURRENT# " gives the name of the dark current file right after the prefix. If more than one DC file is given, then the average of all are used.

    <foreground filename>
The foreground string gives the name of the actual contrast of the sample.

    # <comment>
The string starting with '#' sign, is treated as a comment.

    <empty string>
Empty strings can be used for grouping and have no influence.


-o, --out=PATH
Output result mask.
Defaults to "sino-<input list>-@.tif".
Used when there is a need to describe a stack of output files. In this case the special character represents the number of the output file in the stack. The result mask is a string which forms the names of the processed slices. The mask should (but not must) contain the '@' character which denotes the position where the slice number will be inserted:

    If there is no '@' in the mask, it is inserted together with the '-' prefix right before the file extension (if any) or at the end of the mask (if no extension).

    If there are more than one '@' character in the mask, only the last of them is replaced by the slice number.
For example the mask

produces file names similar to this (for the 13th slice):

Note that the path(s) to the files must exist: the program will not create any directory.
-C, --contrast=STRING
Type of the contrast component.
The component of the contrast to extract via the (E)DEI method and then reconstruct. Must be one of the following strings (case insensitive):
A, ABS, ABSORPTION - for the absorption contrast
P, PHS, PHASE - for the phase contrast
-z, --distance=FLOAT
Object-to-detector distance (mm)
More correctly the distance from the contact print plane and the detector plane where the image given by the argument <far-intensity> was taken. The value is required only to produce the physically correct output. If not set the results are qualitative rather than quantitative.
-r, --resolution=FLOAT
Pixel size of the detector (micron)
The value is required only to produce the physically correct output. If not set the results are qualitative rather than quantitative.
-l, --alpha=FLOAT
The alpha-parameter of the MBA.
-w, --wavelength=FLOAT
Wavelength of the X-Ray (Angstrom)
Only needed together with -l, --alpha.
-g, --gamma=FLOAT
Gamma coefficient of the BAC.
Must be a value around 1.0 (theoretical).
-s, --slice=STRING
Slices to be processed.
The string describing the slices in the image stack which are to be processed. Must be a string consisting only of numbers, commas ',' minus signs '-' and the character 'n'. First the string is divided into the substrings separated by comas and each substring is processed on it's own:

    The string consisting only of numbers is read as the integer and added to the list of the reconstructed slices.

    The string with the minus sign surrounded by the numbers adds all slices in the range into the list.

    If the string starts with the minus then the range start is assumed to be 1.

    If minus is the last character in the string then the range finishes at the maximum slice number.

    If the string has negation prefix 'n' then the slice(s) are excluded from the previously formed list.

    If all substrings have 'n' prefix or the first substring contains only it, then the meaning of the whole string is "all except ...".

    Two and more negations are interpreted as a single one.

    If no slice string is given then all slices are reconstructed.
For example the following string:

requests processing of the slices with numbers 1, 2, 4, 6, 9, 20 to 400, 440 to 449, 471 to 500, 800 to 909, 915 and 921 to the end.
-c, --center=FLOAT[:FLOAT]
Variable rotation center.
Deviation of the rotation axis from the center of the sinogram.
Allows you to erase some minor artifacts in the reconstructed image which come from the inaccurate rotation position. The two float values are the coefficient "bconf" and "aconf" respectively for the calculation of the deviation of the rotation center as

    center = bconf + aconf * slice
with the "slice" being the index in the projection image. If only first of two floats is given then aconf=0.0 (i.e. the rotation axis is strictly vertical).
Note: in most real CT or TS experiments the rotation center of the sample is not exactly in the center of the recoded projection. Moreover, the rotation axis may be inclined from the vertical orientation. If the inclination of the rotation axis is noticeable for your desired spatial resolution then you should rotate the images before the actual reconstruction. You can do it in the batch using "mogrify" command from the ImageMagick package with the -rotate option.
-f, --filter=STRING[:FLOAT]
Filtering window used in the CT.
Defaults to "RAMP".
Must be a string consisting of two parts (the second one is optional): the filter name and float parameter of the filter. The filter name can be one of the following (case insensitive):










Optional float value is required only if the filter is KAISER or GAUSS and represents the lpha parameter of the KAISER filter or igma parameter of the GAUSS filter. Description of the filtering functions and their graphs can be found in the html documentation.
-t, --threads=UINT
Number of threads used in calculations.
If the option is not used the optimal number is calculated automatically.
-i, --int
Output image(s) as integer.
If this option is not set, the output format defaults to the 32-bit float-point TIFF (regardless of the extension). If it is set, the image format is derived from the output file extension (TIFF if the extension does not correspond to any format).

Standard options.

-v, --verbose
Verbose output.
-?, --usage
Outputs brief usage message.
-h, --help
Outputs help message.
When combined with the "-v|--verbose" option may output more detailed message.



ctas(1), ctas-bg(1), ctas-ct(1), ctas-ct-abs(1), ctas-ct-dei(1), ctas-ct-edei(1), ctas-ct-ipc(1), ctas-dei(1), ctas-edei(1), ctas-ipc(1), ctas-f2i(1), ctas-ff(1), ctas-sino(1), ctas-sino-abs(1), ctas-sino-dei(1), ctas-sino-ipc(1), ctas-ts(1), ctas-ct-line(1)



Standard options.

This document was created by man2html, using the manual pages.
Time: 06:49:20 GMT, July 29, 2010