Computational Complexity Optimization of Vehicle Video-Recognition Algorithm Using the Virtual Detectors Method

The paper presents the research of computational complexity optimization of a virtual detector method for video-based vehicle recognition. This method enables a more flexible and adaptive analytical research and optimizes the resource utilization. In addition, the optimization of the computational complexity of the video recognition algorithm required to improve accuracy and speed is investigated. This approach allows to significantly reduce the load on computational resources and ensure the accuracy and efficiency of vehicle recognition even when the frame rate is reduced. The paper presents a system for monitoring the dynamics of traffic flows based on the method of virtual detectors - «ViDeS» (Virtual Detectors System) for which the analysis of recognition efficiency depending on the frame rate is conducted. The developed software allows to increase the efficiency of traffic flow vehicle recognition