%
% Complete documentation on the extended LaTeX markup used for Insight
% documentation is available in ``Documenting Insight'', which is part
% of the standard documentation for Insight.  It may be found online
% at:
%
%     http://www.itk.org/

\documentclass{InsightArticle}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%
%  hyperref should be the last package to be loaded.
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\usepackage[dvips,
bookmarks,
bookmarksopen,
backref,
colorlinks,linkcolor={blue},citecolor={blue},urlcolor={blue},
]{hyperref}
% to be able to use options in graphics
\usepackage{graphicx}
% for pseudo code
\usepackage{listings}
% subfigures
\usepackage{subfigure}


%  This is a template for Papers to the Insight Journal. 
%  It is comparable to a technical report format.

% The title should be descriptive enough for people to be able to find
% the relevant document. 
\title{Robust Automatic Threshold Selection}

% Increment the release number whenever significant changes are made.
% The author and/or editor can define 'significant' however they like.
% \release{0.00}

% At minimum, give your name and an email address.  You can include a
% snail-mail address if you like.
\author{Ga\"etan Lehmann}
\authoraddress{INRA, UMR 1198; ENVA; CNRS, FRE 2857, Biologie du D\'eveloppement et
Reproduction, Jouy en Josas, F-78350, France}

\begin{document}
\maketitle

\ifhtml
\chapter*{Front Matter\label{front}}
\fi


\begin{abstract}
\noindent
Robust Automatic Threshold Selection (RATS) \cite{Kitler1985} is a fast and noise robust
automatic thresholding method based on gradients in the image. The basic idea is to choose
the threshold at the intensity where the gradient are the highest.
\end{abstract}

% \tableofcontents

\section{Description}

RATS is a method to automatically compute a threshold without using the histogram of
the image. It compute a weighted average of the image intensity, using the gradient
of the image, as shown in the following equation:

$$
T = \frac{\sum\limits_{p~in~D}G(p)~.~I(p)}{\sum\limits_{p~in~D}G(p)}
$$

where $I$ is the input image, and $G$ is the gradient of $I$. Several methods may
be used to compute $G$. Both $I$ and $G$ are defined on the domain $D$.

As detailed in \cite{Wilkison98}, using the same method with a pow of the gradient
gives better results with noisy images, by giving a higher weight to the high gradients:

$$
T = \frac{\sum\limits_{p~in~D}G(p)^m~.~I(p)}{\sum\limits_{p~in~D}G(p)^m}
$$

with $m$ greater than $1$, typically $2$.



\section{Implementation}

The filter provided in this article use a pow of 1 by default, to fit the historical algorithm.
Because there is many ways to compute the gradient, the user is in charge to compute the gradient
magnitude image, and to pass it to the filter.
A mask can also be used to limit the computation to a given zone in the image.

\section{Code example}

\small \begin{verbatim}
#include "itkImageFileReader.h"
#include "itkImageFileWriter.h"
#include "itkSimpleFilterWatcher.h"

#include "itkRobustAutomaticThresholdImageFilter.h"
#include "itkGradientMagnitudeRecursiveGaussianImageFilter.h"

int main(int argc, char * argv[])
{

  if( argc != 4 )
    {
    std::cerr << "usage: " << argv[0] << " inputImage outputImage pow" << std::endl;
    exit(1);
    }

  const int dim = 2;
  
  typedef unsigned short PType;
  typedef itk::Image< PType, dim > IType;

  typedef float RPType;
  typedef itk::Image< RPType, dim > RIType;

  typedef itk::ImageFileReader< IType > ReaderType;
  ReaderType::Pointer reader = ReaderType::New();
  reader->SetFileName( argv[1] );
  
  typedef itk::GradientMagnitudeRecursiveGaussianImageFilter< IType, RIType > GradientType;
  GradientType::Pointer gradient = GradientType::New();
  gradient->SetInput( reader->GetOutput() );
  gradient->SetSigma( 10 );
  gradient->Update();

  typedef itk::RobustAutomaticThresholdImageFilter< IType, RIType > FilterType;
  FilterType::Pointer filter = FilterType::New();
  filter->SetInput( reader->GetOutput() );
  filter->SetGradientImage( gradient->GetOutput() );
  filter->SetPow( atof(argv[3]) );

  itk::SimpleFilterWatcher watcher(filter, "filter");

  typedef itk::ImageFileWriter< IType > WriterType;
  WriterType::Pointer writer = WriterType::New();
  writer->SetInput( filter->GetOutput() );
  writer->SetFileName( argv[2] );
  writer->Update();

  return 0;
}
\end{verbatim} \normalsize

% \url{http://www.itk.org}
% \code{Insight/Documentation/Style.pdf}
% \section{Principles of Solar Spot Detection}
% \cite{ITKSoftwareGuide}.
% \doxygen{ImageToImageFilter}

% \small \begin{verbatim}
% \end{verbatim} \normalsize


% The {itemize} environment uses a bullet for each \item.  If you want the 
% \item's numbered, use the {enumerate} environment instead.
% \begin{itemize}
%   \item  Insight Toolkit 2.4.
%   \item  CMake 2.2
% \end{itemize}

% \ref{cthead1}

\begin{figure}[htbp]
\centering
\includegraphics{cthead1}
\caption{The input image.\label{cthead1}}
\end{figure}

\begin{figure}[htbp]
\centering
\includegraphics{test}
\caption{The thresholded image.\label{cthead1}}
\end{figure}


\section{Note on access to the cited articles}
None of the two articles cited are available easily (for free), and the content of
this contribution is based on other sources which are using those article.
You may found some difference in the algorithm compared to the original ones.

A day will come when there will be nothing else than open access...

\appendix



\bibliographystyle{plain}
\bibliography{InsightJournal}
% \nocite{ITKSoftwareGuide}

\end{document}