Video-Based Vehicle Counting Framework

Abstract

The continuous development in the construction of transportation infrastructure has brought enormous pressure to traffic control. Accurate and detailed traffic flow information is valuable for an effective traffic control strategy. This paper proposes a video-based vehicle counting framework using a three-component process of object detection, object tracking, and trajectory processing to obtain the traffic flow information. First, a dataset for vehicle object detection (VDD) and a standard dataset for verifying the vehicle counting results (VCD) were established. The object detection was then completed by deep learning with VDD. Using this detection, a matching algorithm was designed to perform multi-object tracking in combination with a traditional tracking method. Trajectories of the moving objects were obtained using this approach. Finally, a trajectory counting algorithm based on encoding is proposed. The vehicles were counted according to the vehicle categories and their moving route to obtain detailed traffic flow information. The results demonstrated that the overall accuracy of our method for vehicle counting can reach more than 90%. The running rate of the proposed framework is 20.7 frames/s on the VCD. Therefore, the proposed vehicle counting framework is capable of acquiring reliable traffic flow information, which is likely applicable to intelligent traffic control and dynamic signal timing. INDEX TERMS Object detection, object tracking, trajectory processing, vehicle counting.

A Lightweight CNN Model for Refining Moving Vehicle Detection From Satellite Videos

“…For example, Ouyang et al [26] applied the YOLOv3 algorithm framework to achieve vehicle targets, obtaining higher accuracy than traditional vehicle detection methods. To obtain the traffic flow information, Dai et al [27] proposed a video-based vehicle counting framework, which applied YOLO and RCNN to detect moving vehicles. These approaches can take advantage of the existing networks that have proven useful and powerful.…”

Object Detection and Tracking Algorithms for Vehicle Counting: A Comparative Analysis

“…Prior to performing counts, their approach classifies the vehicle into different classes based on their distinct features. (Dai et al 2019) deployed a video based vehicle counting technique using a three-step strategy of object detection, tracking and trajectory analysis. Their method uses a trajectory counting algorithm that accurately computes the number of vehicles in their respective categories and tracks vehicle routes to obtain traffic flow information.…”

Technical and Economic Feasibility Assessment of a Cloud-Enabled Traffic Video Analysis Framework

“…In traditional ITS, the loop detector and microwave sensors have been the primary choices to gather traffic information. However, the data collected by these traditional sensors are aggregated on the sensor side so that detailed information is lost, and installing such sensors is also costly (Dai et al 2019). In addition, traditional video surveillance requires watchstanders to pay close attention to hundreds of screens simultaneously, which is challenging and tedious (Liu et al 2013).…”

“…Typically, based on computer vision (CV) techniques, traffic cameras have been considered capable sensors for data extraction and analysis in various ITS applications, such as congestion detection, vehicle counting, and traffic anomaly detection (Chakraborty et al 2018;Dai et al 2019;Kumaran et al 2019). Such ITS approaches mainly consist of vehicle detection and tracking.…”

“…For vehicle tracking, many tracking algorithms have been proposed such as DeepSORT (Wojke et al 2017), IOU (Bochinski et al 2018), KLT (Lucas et al 1981) and DCF (Lukezic et al 2017). Taking advantage of computer vision techniques, Dai et al (2019) have proposed a video-based vehicle-counting framework using YOLOv3 and kernelized correlation filters (KCF) as detector and tracker, respectively, which achieved an 87.6% counting accuracy running at 20.7 fps, and their video real-time rate was 1.3 (by taking the duration of their framework's runtime divided by the duration of their test videos). Also, Meng et al (2020) have proposed a correlation-matched tracking algorithm combined with SSD for vehicle counting which reached 93% counting accuracy running at 25 fps on an expressway dataset.…”