[ctsim.git] / doc / ctsim-textui.tex

\chapter{The Command Line Interface}\label{ctsimtext}\index{ctsimtext}%
\setheader{{\it CHAPTER \thechapter}}{}{}{\ctsimheadtitle}{}{{\it CHAPTER \thechapter}}%
\ctsimfooter%

\ctsimtext\ is a master shell for all of the command-line utilities. The
command-line utilities can perform most of the functions of the graphical
shell. These command-line utilities are especially appropriate for use on
systems without graphical capability, batch files, and parallel processing
with a Beowulf-type computer cluster.

\usage \ctsimtext\ can be invoked via three different
methods.
\begin{enumerate}\itemsep=3pt
\item \ctsimtext\ can executed without any parameters. In that case,
\ctsimtext\ offers a command-line to enter the function-names and
their parameters. The output of the command is displayed. Further
commands may be given to \ctsimtext. The shell is exited by the
\texttt{quit} command. \ctsimtext\ uses the
\urlref{readline}{http://www.gnu.org} library on UNIX and Linux platforms
to provide command history processing.

\item \ctsimtext\ can also be called to
execute a single command. This is especially useful for batch
files containing multiple \ctsimtext\ commands. This is invoked by
calling\\ \texttt{ctsimtext function-name parameters...}.

\item Using operating systems that support soft or hard linking of
files (such as UNIX and Linux), the executable file \ctsimtext\ can
be linked to the function names. This is automatically done by
the installation file for Linux. Thus, to use \ctsimtext\ with the
function name \texttt{pjrec}, the below command can be executed:\\
\texttt{pjrec parameters...} \\
as a shortcut rather than the equivalent command \\
\texttt{ctsimtext pjrec parameters...}

\end{enumerate}

\section{Parallel Processing}
\ctsimtext\ can be used to spread it's processing over a cluster. Specifically,
\ctsimtext\ supports the \urlref{LAM}{http://www.mpi.nd.edu/lam} version of
the MPI environment. On platforms with LAM installed, a parallel version of
\ctsimtext\ is created. The name of this program is \texttt{ctsimtext-lam}.
The functions that take advantage of the parallel processing are
\texttt{phm2if}, \texttt{phm2pj}, \texttt{pjrec}.

This parallel processing version has been tested with excellent results on
a 16-CPU \urlref{Beowulf}{http://www.beowulf.org} cluster.


\section{if1}\label{if1}\index{if1}%
Performs math functions on a single image. The commands works with
both real and complex valued images.

\usage
\texttt{if1 input-filename output-filename [options...]}

\textbf{Options}
\begin{twocollist}
  \twocolitem{\doublehyphen{invert}}{Negate pixel values.}
  \twocolitem{\doublehyphen{log}}{Take natural logrithm of pixel values.}
  \twocolitem{\doublehyphen{exp}}{Take natural exponent of pixel values.}
  \twocolitem{\doublehyphen{sqr}}{Take square of pixel values.}
  \twocolitem{\doublehyphen{sqrt}}{Take square root of pixel values.}
\end{twocollist}

\section{if2}\label{if2}\index{if2}%
Performs math functions on a two images. The command works with both
real and complex valued images.

\usage
\texttt{if2 input-filename1 input-filename2 output-filename [options...]}

\textbf{Options}
\begin{twocollist}
  \twocolitem{\doublehyphen{add}}{Add the two images.}
  \twocolitem{\doublehyphen{sub}}{Subtract the two images.}
  \twocolitem{\doublehyphen{multiply}}{Multiply the two images.}
  \twocolitem{\doublehyphen{divide}}{Divide the two images.}
  \twocolitem{\doublehyphen{comp}}{Statistically compare the two images.}
  \twocolitem{\doublehyphen{column-plot n}}{Plot the values of a particular column.}
  \twocolitem{\doublehyphen{row-plot n}}{Plot the values of a particular row.}
\end{twocollist}

\section{ifexport}\label{ifexport}\index{ifexport}%
Export an image file to a standard graphics file.

\usage
\texttt{ifexport input-filename output-filename -\,-format }\emph{graphic-format} \texttt{ [options...]}

\textbf{Options}
\begin{twocollist}
  \twocolitem{\doublehyphen{format}}{
  \begin{itemize}\itemsep=0pt
    \item \texttt{pgm} - Portable graymap format.
    \item \texttt{pgmasc} - ASCII PGM format.
    \item \texttt{png} - Portable network graphics format.
    \item \texttt{png16} - 16-bit PNG format.
  \end{itemize}}
  \twocolitem{\doublehyphen{center}}{Set center of intensity window.
  \begin{itemize}\itemsep=0pt
    \item \texttt{median}
    \item \texttt{mode}
    \item \texttt{mean}
  \end{itemize}}
  \twocolitem{\doublehyphen{auto}}{Set half-width of intensity window as a multiple of the standard deviation.
  \begin{itemize}\itemsep=0pt
     \item \texttt{full}
     \item \texttt{std0.1}
     \item \texttt{std0.5}
     \item \texttt{std1}
     \item \texttt{std2}
     \item \texttt{std3}
  \end{itemize}}
  \twocolitem{\doublehyphen{scale}}{Set size of output image. A value of
  \texttt{1} is default and creates an output image the same size as the input image.}
  \twocolitem{\doublehyphen{min}}{Set the minimum intensity value.}
  \twocolitem{\doublehyphen{max}}{Set the maximum intensity value.}
\end{twocollist}

\section{ifinfo}\label{ifinfo}\index{ifinfo}%

Displays information about an imagefile.

\usage
\texttt{ifinfo input-filename [options...]}

\textbf{Options}
\begin{twocollist}
  \twocolitem{\doublehyphen{labels}}{Display history labels.}
  \twocolitem{\doublehyphen{no-labels}}{Suppress history labels.}
  \twocolitem{\doublehyphen{stats}}{Display image statistics.}
  \twocolitem{\doublehyphen{no-stats}}{Suppress image statistics.}
\end{twocollist}

\section{phm2pj}\label{phm2pj}\index{phm2pj}%
Simulates collection of X-rays data (projections) around a phantom object.

\usage
\texttt{phm2pj projection-filename number-detectors number-views [options...]}

\begin{twocollist}
\twocolitem{\doublehyphen{phantom}}{Select a standard phantom.
\begin{itemize}\itemsep=0pt
\item \texttt{herman}
\item \texttt{shepp-logan}
\item \texttt{unit-pulse}
\end{itemize}
}
\twocolitem{\doublehyphen{phmfile}}{Reads a user-created phantom file.}

\twocolitem{\doublehyphen{geometry}}{Sets the scanner geometry. Valid values are:
  \begin{itemize}\itemsep=0pt
    \item \texttt{parallel}
    \item \texttt{equiangular}
    \item \texttt{equilinear}
  \end{itemize}
}

\twocolitem{\doublehyphen{nray}}{ Number of samples per each detector}

\twocolitem{\doublehyphen{rotangle}}{The rotation angle as a multiple of pi.
For parallel geometries use a rotation angle of \texttt{1} and for equilinear and equiangular
geometries use a rotation angle of \texttt{2}. The default is to use to
appropriate rotation angle based on the geometry.}

\twocolitem{\doublehyphen{view-ratio}}{Sets the field of view as a ratio of the diameter of the phantom.
    For normal scanning, the default value of \texttt{1.0} is optimal.}

\twocolitem{\doublehyphen{scan-ratio}}{Sets the length of scanning as a ratio of the view diameter.
    For normal scanning, the default value of \texttt{1.0} is optimal.}

\twocolitem{\doublehyphen{focal-length}}{Sets the distance between the radiation source
 and the center of the object as a ratio of the radius of the object.
   For parallel geometries, a value of \texttt{1.0} is optimal. For other
    geometries, this should be at least \texttt{2.0} to avoid artifacts.}
\end{twocollist}


\section{phm2if}\label{phm2if}\index{phm2if}%
Converts a geometric phantom object into an imagefile.

\usage
\texttt{phm2if phantom-filename image-filename [options...]}

\textbf{Options}
\begin{twocollist}
  \twocolitem{\doublehyphen{nsamples}}{Number of samples in x \& y directions per pixel}
  \twocolitem{\doublehyphen{view-ratio}}{Sets the view ration. For normal scanning,
  the default value of \texttt{1.0} is optimal.}
\end{twocollist}

\section{pj2if}\label{pj2if}\index{pj2if}%
Convert a projection file into an imagefile.

\usage
\texttt{pj2if projection-filename image-filename x-size y-size [options...]}

\textbf{Options}
\begin{twocollist}
\twocolitem{\doublehyphen{dump}}{Print all projection data to the console.}
\end{twocollist}

\section{pjinfo}\label{pjinfo}\index{pjinfo}%
Displays information about a projection file.

\usage
\texttt{pjinfo projection-filename [options...]}

\textbf{Options}
\begin{twocollist}
  \twocolitem{\doublehyphen{binaryheader}}{Dump the binary header to the standard output.
  This option is only used when manually creating a composite projection file from
  several different projection files.}
  \twocolitem{\doublehyphen{binaryview}}{Dump binary view data to the standard output.
  This option is only used when manually creating a composite projection file from
  several different projection files.}
  \twocolitem{\doublehyphen{startview}}{Sets starting view to display. Default is \texttt{0}.}
  \twocolitem{\doublehyphen{endview}}{Sets ending view to display. Default is the last view.}
  \twocolitem{\doublehyphen{dump}}{Print all projection data to the console.}
\end{twocollist}

\section{pjrec}\label{pjrec}\index{pjrec}%
Reconstructs the interior of an object from a projection file.

\usage
\texttt{pjrec projection-filename image-filename image-cols image-rows [options...]}

\textbf{Options}

\begin{twocollist}
\twocolitemruled{\textbf{Parameter}}{\textbf{Options}}
\twocolitem{\doublehyphen{filter}}{Selects which filter to apply to
each projection. To properly reconstruct an image, this filter
should be multiplied by the absolute value of distance from zero
frequency.
\begin{itemize}\itemsep=0pt
\item \texttt{abs\_bandlimit}
\item \texttt{abs\_cosine}
\item \texttt{abs\_hamming}
\end{itemize}
} \twocolitem{\doublehyphen{filter-parameter}}{Sets the alpha level
for Hamming window. At setting of \texttt{0.54}, this equals
the Hanning window.}

\twocolitem{\doublehyphen{filter-method}}{Selects the filtering
method. For large numbers of detectors, \texttt{rfftw} is optimal.
For smaller numbers of detectors, \texttt{convolution} might be a
bit faster.
\begin{itemize}\itemsep=0pt
\item \texttt{convolution}
\item \texttt{fourier} - Uses simple Fourier transform.
\item \texttt{fourier-table} - Optimizes Fourier transform by precalculating trigometric functions.
\item \texttt{fftw} - Uses complex-valued Fourier transform with the \emph{fftw} library.
\item \texttt{rfftw} - Uses optimized real/half-complex Fourier transform.
\end{itemize}
}
\end{twocollist}

\begin{twocollist}
\twocolitem{\doublehyphen{filter-generation}}{Selects the filter
generation. With convolution, \texttt{direct} is the proper method
to select. With any of the frequency methods,
\texttt{inverse-fourier} is the best method.
\begin{itemize}\itemsep=0pt
\item \texttt{direct}
\item \texttt{inverse-fourier}
\end{itemize}
}

\twocolitem{\doublehyphen{interpolation}}{Interpolation technique.
\texttt{cubic} is optimal when the
data is smooth. Smooth data is obtained by taking many projections and/or
using a smoothing filter. In the absence of smooth data, \texttt{linear} gives better results and
is many times faster than cubic interpolation.

\begin{itemize}\itemsep=0pt
\item \texttt{nearest}
\item \texttt{linear}
\item \texttt{cubic}
\end{itemize}
}

\twocolitem{\doublehyphen{backprojection}}{Selects the
backprojection technique. A setting of \texttt{idiff} is optimal.
\begin{itemize}\itemsep=0pt
\item \texttt{trig} - Use trigometric functions at each image point.
\item \texttt{table} - Use precalculated trigometric tables.
\item \texttt{diff} - Use difference method to iterate within image.
\item \texttt{idiff} - Use integer iteration math.
\end{itemize}
}

\twocolitem{\doublehyphen{zeropad}}{Zeropad factor. A setting of
\texttt{1} is optimal whereas a zeropad of \texttt{0} performs no zeropadding.}

\end{twocollist}