HazGTSimulator

## An introduction to the files. ### 1. Dataset * UniformSampling: the result data of uniform sampling experiment. * UniformSampling/Scene*/Hazard/UniformSampling.png: the curves of recall and precision. * UniformSampling/Scene*/Hazard/WrongLabelsNumber.png: the curve of the number of perception errors. * UniformSampling/Scene*/Hazard/Hazard_Value/GroundtruthLabels: the detection files of the groundtruth of a simulation. * UniformSampling/Scene*/Hazard/Hazard_Value/labels: the detection files of yolo model. * UniformSampling/Scene*/Hazard/Hazard_Value/flabels: the detection files of frcnn model. * UniformSampling/Scene*/Hazard/Hazard_Value/Source: the images captured in a simulation. * UniformSampling/Scene*/Hazard/Hazard_Value/Paint: the images with groundtruth labels. * UniformSampling/Scene*/Hazard/Hazard_Value/Yolov7: the images detected by yolov7. * UniformSampling/Scene*/Hazard/Hazard_Value/FRCNN: the images detected by frcnn. * UniformSampling/Scene*/Hazard/Hazard_Value/Hazards.json: the attributes file used in a simulation. * Optimization: the result data of optimization. * Optimization/Yolo: the result data of optimization using yolo as perception model. * Optimization/Frcnn: the result data of optimization using frcnn as perception model. * Optimization/F1SUM.pdf: the outcome picture about F1-type failures. * Optimization/F2SUM.pdf: the outcome picture about F2-type failures. * Optimization/*/*/GA.json: the result data of optimization using GA. * Optimization/*/*/RA.json: the result data of optimization using random algorithm. * Optimization/*/*/NH.json: the result data of control group in whose scenario there is no hazard. <br> ### 2. Tool * HazGTSimulator: the simulator we used to generate traffic videos and images * Click HazGTSimulator.exe. * Use the toggles or sliders located at the top left of the screen to set the scenario. Alternatively, use the panel located at the bottom left of the screen, where you can choose to import settings using files in three formats: hazs, hazgt, and json. * Click the button Start Simulation which is located at the bottom of the screen. Before you start a simulation, make sure you know where the output files are, which can be modified through the panel mentioned at RQ2. * Click the button Uniform Sampling and select an output folder with enough space.