VCTSELECTVisionX V4
NAME
vctselect − extract simple intensity features from CT images
SYNOPSISvctselect [if=infile] [of=ofile] [th=] [lo=] [hi=] [-nd] [-n] [-a] [-v]
DESCRIPTIONvctselect accepts a CT integer image as input and outputs a byte binary-valued output image. The -nd option uses an automatic alogrithm to identify any "no-data" regions of the CT image. This algorithm examins the axial image corners to identify a conistent low intensity "no-data" pixel value that is less than -1000 HU. DIfferent manufactures use different values.
This utility may be used for simple image thresholding and other basic functions. With no options specified a zero valued byte image with the same dimesnions as the input image is created. If -n is specifed then the image output is inverted. If th= is specified then the CT image is thresholded at that value.
CONSTRAINTSThe no-data algortihm assumes a conventional axial slice data orientation. Only short integer images are accepted for input. Slope values other than 1.0 are currently not supported
OPTIONS
if= |
Input file, must be a short interger CT image preferabley with valid DICOM scale information. | ||
of= |
Output image is always a byte binay-valued image | ||
th= |
specify the threshold value for an image threshold operation. The thersold should be specified in Houndsfield units (HU) for valid DICOM images. | ||
lo= |
set the low (false) value for the output image (default 0 ) | ||
hi= |
set the high (true) value for the output image (default 255) | ||
-n |
negate the values of the output image | ||
-a |
absolute thresholding, the threshold value is applied direclty to pixel values (the DICOM intercept value if available is ignored) | ||
-v |
selects the verbose mode and prints some information about its action. |
Compute a binary image of the no-data regions in a CT image
vctselect -nd if=im of=out
Threshold an image at -600 HU
vctselect th=-600 if=im of=out
Create a
zero-valued byte image with the same dimensions
as a CT image
vctselect if=im of=out
AUTHORA. P. Reeves