deletions | additions       diff --git a/section_Conclusion_The_usage_of__.tex b/section_Conclusion_The_usage_of__.tex  index 28d95a1..e604ec6 100644  --- a/section_Conclusion_The_usage_of__.tex  +++ b/section_Conclusion_The_usage_of__.tex 
...
The usage of a genetic algorithm on manually annotated video sequences shows that there is room for significant improvements over the default tracking parameters. Considering the strong impact that winter conditions had on the performance results, a logical next step would be the evaluation of other meteorological conditions such as precipitation, low visibility (fog), nighttime and high winds (affecting camera stability). There is also additional work to be done comparing calibration efforts to traffic data that could not be extracted with proper significance from ten minutes of video. Gap time, time-to-collision and user pairs are examples of such data. Different types of optimization algorithms and performance metrics could also be evaluated based on both computation time and reliability of the solutions.  The real world application is to develop a reliable single or dynamic set of parameters that could be calibrated and applied to both fixed and mobile cameras to be used under all conditions. Combined with automated analysis tools, a wide-scale application of automatically calibrated video tracking could provide researchers with a comprehensive database from which surrogate safety measures could be calculated.  The first portion of this paper finds a strong correlation between traffic counts and the measure of tracking performance (MOTA), as well as a correlation of up to -0.879 in the computation of MOTA for certain parameters using Spearman's rank correlation. The second portion introduces the concept of transferability of sets of parameters optimized on one specific video sequence. Parameters for summer sequences are demonstratively not applicable to winter conditions as results for this sequence did not surpass a MOTA of 0.150 unless specifically optimized for.