This is a desktop app for detecting highway lane violations. Illegal parking, Wrong direction, Illegal over taking can be detected. This can be implemented realtime on CPU thanks to OpenVino model optimisation. To track objects (vehicles), SORT algorithm with Kalmann Filter (tracker) and Hungarian algorithm (data association) has been used. Pre-trained openvino vehicle detection model (MobileNetV2 SSD) has been used for detection. Tkinter library is used for GUI
Demo: https://www.youtube.com/watch?v=VP6AhOG4vOY
Create a new conda environment. If you dont have conda installed download miniconda
conda create -n hwlvd python=3.8 Clone this repository to your computer and navigate to the directory.
Activate new enviroment
conda activate hwlvd Install all the libraries used
pip install -r requirements.txt Then run the model
python run_sort.py
Select input video for monitoring.
Select a directory for saving the output if necessary.
Insert the threshold value for detection model.
Insert the duration which is the limit that the program will allow a vehicle not be considered as parked.
Used for stabilizing the park detection.
Direction the vehicles are supposed to go legally..
Whether to save the video or not.
Load the model and other information (a window will pop out to manually draw the boundaries for lanes).
Play the video.
Stop the video.
Quit the program.
Number of vehicles that are detected as illegally parked at the moment.
Number of vehicles that are detected as violated the lane discipline at the moment.
Number of vehicles that are detected as driving at the wrong direction at the moment.
The amount of time(ms) that has been spent on processing one frame.