Detecting, identifying, and counting vehicles based on deep learning algorithms in video surveillance systems

Detection, identification, and automatic counting of vehicles using video surveillance cameras plays an important role in the field of intelligent transportation management. Despite the progress that researchers have made in these cases, its operational implementation still faces challenges such as "various environmental conditions", "unbalanced data sets", "accuracy" and "speed". Therefore, research can be useful in solving these issues. The proposed algorithm for detection, classification, and counting will be based on deep learning. In this research, after applying the proposed initial preprocessing algorithm, we use the YOLO algorithm to detect and classify vehicles. The DeepSORT algorithm is also used to track several vehicles at the same time. For the accurate counting of vehicles, a developed method is also proposed to increase the processing accuracy. By applying the proposed pre-processing and counting techniques, the practical results show that the "call" criterion in the video with detection at night challenge has been increased to 99.18%.

Detecting, identifying, and counting vehicles based on deep learning algorithms in video surveillance systems Keywords:

Machine Vision , Deep Learning , Vehicle detection , Car Classification , Car Counting