deletions | additions       diff --git a/section_Experimental_Results_subsection_The__.tex b/section_Experimental_Results_subsection_The__.tex  index 0f99ecb..db38df2 100644  --- a/section_Experimental_Results_subsection_The__.tex  +++ b/section_Experimental_Results_subsection_The__.tex 
...
\caption{Counts (left column) and average speeds (right column) measured at the entrance and exits of the merging zone for each video sequence (corresponding to each row)}  \end{figure}  These results were used to validate the choice of MOTA as a measure of performance. As expected, there is a strong correlation between MOTA and the number of objects tracked due to the nature of MOTA. It %It  is important to note that the total vehicle counts are not necessarily comparable to the number of objects in the ground truth since the genetic algorithm is run on the analysis zones which do not include the whole video. Lower MOTA These  results also were used to validate the choice of MOTA as a measure of performance.   As expected, there is a strong correlation between MOTA and the number of road users tracked given the definition of MOTA: if tracking errors are uniformly spread over the analysis zone, since MOTA represents the average percentage of correctly tracked road user instants, one can expect that the resulting counts will be around the true number of road users multiplied by MOTA. MOTA seems therefore to be a good indicator of total counting accuracy. On the other hand, average spot speeds seem relatively insensitive to MOTA, except for lower values of MOTA which  tend to have be associated with  a larger rangein terms  of average tracked speed. extracted spot speed, especially for S1W. This can be related to the counts: lower MOTA values generate low counts and the spot speeds are a small random sample of all road users with high variability.  As the performance increases in lanes that have a significant number of vehicles, the average tracked spot  speeds converge. Therefore, converge (one can still observe a lot of variability independently of MOTA for lane 1 at the exit).   %Therefore,  the video chosen for annotation should have a reasonable flow in every lane of the analysis zone. The two higher resolution cameras camera views  (S3V1 and S3V2) had very few results of poor performance. This suggests a lower sensitivity to the chosen tracking  parameters used in the genetic algorithm. chosen for optimization.  S1W presents a special case where only 49 percent 49~\%  of specimens had any tracked objects. These two cases are discussed in subsequent sections. \subsection{Correlation of Tracking Parameters with Tracking Accuracy} 
...