ABSTRACT
The main focus of this study is on vehicle detection and counting, especially for traffic control. In the field of highway regulators, vehicle detection and counting are becoming increasingly significant. However, because to the diverse structures of cars, detection remains difficult, which has a direct impact on the accuracy of a vehicle count. Based on OpenCV technologies, this study addresses video-based algorithms for vehicle recognition and counting. To find forefront objects in video sequels, the proposed solution employs the background subtraction method. Several OpenCV techniques, such as thresholding, adaptive morphological operations, and hole filling, are then used to improve the accuracy of detecting moving cars. Finally, virtual identification zones are used to count vehicles.
