A Novel Deep Learning Based Method for Detection and Counting of Vehicles in Urban Traffic Surveillance Systems

Abstract

Abstract. In intelligent transportation systems (ITS), it is essential to obtain reliable statistics of the vehicular flow in order to create urban traffic management strategies. These systems have benefited from the increase in computational resources and the improvement of image processing methods, especially in object detection based on deep learning. This paper proposes a method for vehicle counting composed of three stages: object detection, tracking and trajectory processing. In order to select the detection model with the best trade-off between accuracy and speed, the following one-stage detection models were compared: SSD512, CenterNet, Efficiedet-D0 and YOLO family models (v2, v3 and v4). Experimental results conducted on the benchmark dataset show that the best rates among the detection models were obtained using YOLOv4 with mAP = 87% and a processing speed of 18 FPS. On the other hand, the accuracy obtained in the proposed counting method was 94% with a real-time processing rate lower than 1.9.

Deep Neural Network-based Approach for Accurate Vehicle Counting

“…Even though these strategies are straightforward to grasp, they struggle in scenarios containing occlusions. Furthermore, detection algorithms encounter numerous problems, particularly in the task of vehicle countings, for example, changes in occlusions, perspectives, illumination effects, and many others [8]. The author in [9] used image keys and interest points to create the scale-invariant feature transform (SIFT) and the speeded-up robust features (SURF).…”