deletions | additions
diff --git a/section_Methodology_The_approach_proposed__.tex b/section_Methodology_The_approach_proposed__.tex
index 4e0f218..a4eade4 100644
--- a/section_Methodology_The_approach_proposed__.tex
+++ b/section_Methodology_The_approach_proposed__.tex
...
\section{Methodology}
The approach proposed in this paper consists in identifying different conditions that may have an impact on tracking performance and
on the accuracy of traffic variables such as counts. For each condition, we need two video samples or regions in the same video where the only or main difference is the change in that condition. The method relies on the following four main steps:
\begin{enumerate}
\item Selection of sites and analysis zones:
Five five different camera views are specifically selected to allow for analysis based on
the chosen
comparison points. conditions to be compared. Ten minutes of video are manually annotated for
each an analysis zone
in each camera view to be used as a baseline for the analysis.
\item Optimization of tracking parameters over
the whole annotated period (10~min): a
ten minute period: The subset of the tracking parameters
used in computer vision are optimized
and saved for each camera view using the chosen measure of
performance for each camera. performance. These results are used to evaluate traffic data and
correlation correlations of parameters
compared to with the measure of performance.
\item Optimization of tracking parameters over the first five minute period:
The best results from the tracking parameters optimized for the first five
annotated minutes of each camera
view are
checked for over-fitting by using applied to the
second whole 10 minute annotated
segment video, as
a separate dataset. Close and far well as to sub-regions of the analysis zones
are tested. (two sub-regions, one near and one far from the camera), in order to evaluate over-fitting.
\item The
best-fit optimized tracking parameters from step 3 are applied to
to the full ten minutes of each
analysis zone camera view to
check for evaluate the transferability between sites, camera types and weather conditions.
\end{enumerate}
The overview of the methodology is presented in FIGURE~\ref{fig:optimization_overview}.
\subsection{Ground Truth Inventory}
The ground truth data is
built using obtained through manual annotation of the source video data using the Urban Tracker
annotation application \cite{Jodoin_2014}.
For each camera sequence 10 minutes of video are manually annotated. There are
5 five video sequences selected from
3 three different roundabouts presented in
Table~\ref{tab:gt_inv}. Table~\ref{tab:gt_inv} (a video sequence correspond to a camera views and the terms are used interchangeably). S1S and S1W were recorded with similar field of views on the first site to compare the weather condition. S1S and S2 show comparable views of two different roundabouts (sites 1 and 2). S3V1 was recorded using the same camera as on S1 and S2, and can be compared to S1S and S2 to evaluate the impact of the resolution. S3V1 and S3V2 allow to compare the impact of the type of camera on the same site, with two different views. For each camera view, an analysis zone covering a merging zone of the roundabout is defined inside the zone where automated tracking is performed (site analysis mask) as can be seen in FIGURE~\label{fig:analysis_zone_trajectories}. All vehicles going through the analysis zone of each camera view were manually tracked for 10~min (with bounding boxes drawn around each vehicle every 5-10 frames). The annotations require between
30 minutes 30~min and one hour per minute of
annotated video, depending on the frame rate and
the traffic flow.
\begin{table}
\caption{Ground Truth
Inventory} Inventory: S1, S2 and S3 refer to the sites 1 to 3, S1S and S1W refer respectively to the videos recorded on the first site in Summer and Winter, and S3V1 and S3V2 refer respectively to the videos recorded simultaneously on the third site with two different cameras covering complementary zones of the roundabout}
\label{tab:gt_inv}
\begin{tabular}{lp{3.2cm}p{2.5cm}p{3cm}ll}
\hline
\textbf{Site} &
\textbf{Date \textbf{Time \&
Time} Date} & \textbf{Number of Annotated Road Users} & \textbf{Camera Type} & \textbf{Conditions} & \textbf{Sample View} \\
\hline
S1S & 12:00pm, July 2012 (Thursday) & 266 & IP Camera 800x600 & Sunny, shadows & \includegraphics[width=4cm]{figures/image2.png}\\
S1W & 8:00am, February 2013 (Friday) & 209 & IP Camera 800x600& Low visibility, winter & \\
...