Combined some figures and got a little farther.

docs/book/getting_started.tex

@@ -67,7 +67,7 @@ \section{Selecting images to process}
 extra work and setup which are not covered in this document, and
 are considered `advanced usage.'  The image type that is easiest
 for the user to provide and for the Stereo Pipeline's \texttt{stereo}
-program to ingest are ISIS 3 \texttt{.cub} files.
+program to ingest are ISIS 3 cube files (\texttt{.cub}).
 
 
 \section{Example data set}
@@ -76,7 +76,6 @@ \section{Example data set}
 pair of Mars Orbiter Camera (MOC)
 \citep{1992JGR....97.7699M,2001JGR...10623429M} images whose Planetary
 Data System (PDS) Product IDs are M01/00115 and E02/01461.
-
 These data can be downloaded from the PDS directly, or they can be found in 
 the \texttt{data/MOC/} directory of your Stere Pipeline distribution.
 
@@ -99,19 +98,22 @@ \section{Example data set}
 the \texttt{Mapping} parameter to `\texttt{NO}' the resultant file
 will be an ISIS cube file that is calibrated, but not map-projected.
 
-\begin{figure}
-\begin{center}
-\includegraphics[height=4.5in]{images/p19-images.png}
+\begin{figure}[b!]
+\begin{minipage}{5.2in}
+\includegraphics[height=3.7in]{images/p19-images.png}
 \hfill
-\includegraphics[height=4.5in]{images/p19-images_zoom.png}
+\includegraphics[height=3.7in]{images/p19-images_zoom.png}
+\end{minipage}
+\hfill
+\begin{minipage}{1.3in}
 \caption[P19 images open in qview zoomed in]{
     \label{p19-images}
     This shows \texttt{E0201461.cub} and \texttt{M0100115.cub} open in
 	ISIS's qview program.  The view on the left shows their full extents
 	at the same zoom level, showing how they have different ground scales.
 	The view on the right shows them both zoomed in on the same feature.
     }
-\end{center}
+\end{minipage}
 \end{figure}
 
 % \section{Examples of Use}
@@ -198,68 +200,92 @@ \section{Tutorial}
     E0201461.cub   M0100115.cub
     > mkdir results
 \end{verbatim}
-
+\noindent
 The \texttt{stereo} program requires a \texttt{stereo.default} file
 which can be altered for your needs.  Its contents are detailed on
 page \pageref{stereo.default}.  You may find it useful to save
 multiple versions of the \texttt{stereo.default} file for various
 processing needs. If you need to do that, be sure to specify which
 configuration file \texttt{stereo} should use with the \texttt{-s}
 option.  If this option is not given, the \texttt{stereo} program
-will search for a file named \texttt{stereo.default} and will
-complain if there isn't one.
+will search for a file named \texttt{stereo.default} in the current
+directory and will complain if there isn't one.
 
-Then run \texttt{stereo} like this:
+Then run \texttt{stereo} like this (there should be a
+\texttt{stereo.default} file distributed along with the example
+data set):
 
 \begin{verbatim}
-    stereo E0201461.cub M0100115.cub results/p19
+    stereo E0201461.cub M0100115.cub results/E0201461-M0100115
 \end{verbatim}
-
+\noindent
 So that last text can be anything you want it to be.  It designates
 the text that \texttt{stereo} will use as a prefix for its many
 output files.  Since the first part is \texttt{results/} this causes
 the program to put the results in that directory with files whose
-names start with \texttt{p19}.  If instead that last
-text was just \texttt{p19} it would have created a
-bunch of files that start with \texttt{p19} in the
+names start with \texttt{E0201461-M0100115}.  If instead that last
+text was just \texttt{E0201461-M0100115} it would have created a
+bunch of files that start with \texttt{E0201461-M0100115} in the
 same directory as the input files.
 
 The \texttt{stereo} program's processing moves through several
 stages which are detailed on page \pageref{entrypoints}.  However,
-once the \texttt{stereo} program completes, it creates a number of 
-files.
+once the \texttt{stereo} program completes, it creates a number of
+files.  A quick look at some of the TIFF files created, can quickly
+give you an idea of what the \texttt{stereo} program did (figure
+\ref{p19-stereo-output}).
 
 
 \begin{figure}
 \begin{center}
-\includegraphics[width=3in]{images/p19-masks.png}
-\caption[P19 images masks]{
-    \label{p19-masks}
-	The masks of pixels which are useful for alignment.
+\includegraphics[width=5in]{images/p19-stereo-output.png}
+\caption[P19 stereo output images]{
+    \label{p19-stereo-output}
+	These are the four viewable \texttt{.tif} files created by
+	the \texttt{stereo} program.  The left two are the aligned
+	images (\texttt{E0201461-M0100115-L.tif} and
+	\texttt{E0201461-M0100115-R.tif}).  The next two images are
+	the mask images (\texttt{E0201461-M0100115-lMask.tif} and
+	\texttt{E0201461-M0100115-rMask.tif}), which indicate which
+	pixels in the aligned images are good to use for the next
+	step.  The image on the right is the Good Pixel map
+	(\texttt{E0201461-M0100115-GoodPixelMap.tif}), which indicates
+	the pixels in grey which were successfully matched with the
+	correlator.  The red pixels were not.  Those red pixels which 
+	are not black in both mask images are optionally filled later
+	during the hole-filling step.
     }
 \end{center}
 \end{figure}
 
-\begin{figure}
-\begin{center}
-\includegraphics[height=8in]{images/p19-goodpixel.png}
-\caption[P19 good pixel image]{
-    \label{p19-goodpixel}
-	The Good Pixel map. 
-	Red pixels are not useful for alignment. 
-    }
-\end{center}
-\end{figure}
-
-\begin{figure}
-\begin{center}
-\includegraphics[width=3in]{images/p19-aligned.png}
-\caption[P19 aligned image]{
-    \label{p19-aligned}
-	The left and right aligned images.
-    }
-\end{center}
-\end{figure}
+% \begin{figure}
+% \begin{center}
+% \includegraphics[height=8in]{images/p19-goodpixel.png}
+% \caption[P19 good pixel image]{
+%     \label{p19-goodpixel}
+% 	The Good Pixel map. 
+% 	Red pixels are not useful for alignment. 
+%     }
+% \end{center}
+% \end{figure}
+% 
+% \begin{figure}
+% \begin{center}
+% \includegraphics[width=3in]{images/p19-aligned.png}
+% \caption[P19 aligned image]{
+%     \label{p19-aligned}
+% 	The left and right aligned images.
+%     }
+% \end{center}
+% \end{figure}
+
+If the above TIFF files look okay, you can probably just go on to
+making a mesh or a DTM from the point cloud file
+(\texttt{E0201461-M0100115-PC.tif}).  However, to get an idea of the
+disparity information that the \texttt{stereo} program created and then
+used to build the point cloud, it can be useful to take a look at that
+disparity information.  The \texttt{stereo} program records this information
+in the \texttt{E0201461-M0100115-D.exr} file.
 
 To begin an examination of the results, you must first examine the
 disparity images. These images show the horizontal and vertical

docs/book/programs.tex

@@ -45,6 +45,15 @@ \section{stereo}
 
 \begin{description}
 
+\item[.vwip]
+One of these Vision Workbench Interest Point files will be created
+for each input image (if your images are \texttt{left.cub} and \texttt{right.cub} you'll get \texttt{left.vwip} and \texttt{right.vwip}).  \emph{MJB: what good are these, what are they, what do they do for us?}
+
+\item[.match]
+This match point file \emph{does what exactly?}  Again, if your
+images are \texttt{left.cub} and \texttt{right.cub} you'll get a
+\texttt{left\_\_right.match} file.
+
 \item[out-D.exr] 
 This is the unfiltered disparity map, straight out of the stereo correlator.
 
@@ -72,6 +81,9 @@ \section{stereo}
 \item[out-lMask.tif] Image mask for the left image.
 \item[out-rMask.tif] Image mask for the right image.
 
+\item[out-stereo.default] This is a copy of the stereo.default file used to during the run of \texttt{stereo} that built all of these files.
+
+
 \end{description}
 
 \emph{MJB: also talk about the .vwiw and the .match files that get generated.}
@@ -118,18 +130,22 @@ \subsection{Entry Points}
 \emph{MJB: discuss the different phases, and how to know when you're
 in which, etc.}
 
-Stage 0 of the processing normalizes the two images and aligns them
+Stage 0 (Preprocessing) normalizes the two images and aligns them
 (thus non-projected images are easier to work with) by locating
 interest points and matching them in both images. The program is
-designed to reject outlying interest points.
+designed to reject outlying interest points.  This stage writes out
+the pre-aligned images and the image masks.
 
-Stage 1 of the processing is correlation and the building of a disparity map. 
+Stage 1 (Correlation) performs the image correlation and the building
+of a disparity map.
 
-Stage 2 of the processing is refinement. 
+Stage 2 (Refinement) performs sub-pixel correlation which further refines the solution.
 
-Stage 3 is filtering and the creating of a ``good pixel'' map. 
+Stage 3 (Filtering) performs filtering of the disparity map and
+creates a ``good pixel'' map.  If enabled, this is also the step
+where holes are filled in.
 
-Stage 4 is triangulation. The disparity map is processed to remove
+Stage 4 (Triangulation). The disparity map is processed to remove
 the effects of interest point alignment and a 3D point cloud is
 created.