Bhattacharyya distance-based tracking: A vehicle counting application

Transportation plays an important role in urban development However, the vehicle growth in Indonesia is not supported by the number of road. Due to this fact, traffic congestion is easily occurred, especially in big cities. Intelligent Transportation System (ITS) has huge contribution to decrease the traffic congestion. In ITS, vehicle detection is one of challenging issue for traffic surveillance. In this paper, adaptive selective background learning and hole filling algorithm are applied to improve the vehicle detection. The validity of the proposed method is tested by using three scenarios and two parameters. The scenarios are bad weather close range (BW-CR), normal weather close range (NW-CR) and normal weather wide range (NW-WR). While, the parameters are the time duration of stopped vehicle detection and the pixel accuracy. Then, the proposed method (Adaptive Selective Background Learning-Hole Filling algorithm) is compared by another previous vehicle detection method. Generally, the result shows that the proposed method yields a significant improvement in vehicle detection. ASBL-HF can detect the stopped and moved vehicle with free noises. Moreover, ASBL-HF has the best accuracy. The accuracy value is about 98.2%.

Vehicle counting is a process to estimate traffic density on roads to assess the traffic conditions for intelligent transportation systems (ITS). Real-time traffic management systems have become popular recently due to the availability of high end cameras and technology. The present traffic management systems focus on speed detection, signal jumping, zebra crossing but not on traffic density estimation. Proposed video-based vehicle counting and tracking method using a video captured on CCTV and handheld mobile cameras. The system can be used in smart cities to create smart traffic light signals, in which duration of each signal depends on real time vehicle density in a particular lane of road. Vehicle counting is performed in two steps: the captured video is sent to You Only Look Once (YOLO) based deep learning framework to detect, count and classify the vehicles. Multi vehicular tracking is adopted using Deep SORTalgorithm to track the vehicles in video frames. Model was trained for six different classes, using Google Colaboratory. Datasets of vehicles specifically pertaining to Indian roads environment is considered for implementation. The performance of the model was analyzed, proposed model has tested and obtained an average counting accuracy of 86.56% while the average precision is 93.85%. The model can be implemented for ascertaining the traffic density on roads and this provides knowledge for infrastructural development to authorities. It can also be an integral part of smart city projects to develop intelligent and smart traffic surveillance system.

Intelligent Transportation Systems (ITS) are increasingly shifting toward data-driven and artificial intelligence-based frameworks that integrate sensing, prediction, and control to enhance mobility efficiency, safety, and adaptability. This dissertation contributes to the development of next-generation ITS through three complementary studies focusing on flexible traffic monitoring, real-time traffic forecasting, and stability-aware vehicle control. The first study develops a multi-drone framework for wide-view corridor traffic analysis and origin-destination (OD) estimation. By coordinating multiple unmanned aerial vehicles, as demonstrated with two drones in this study, the framework enables continuous vehicle tracking across extended corridors. A deep learning-based pipeline combining vehicle detection and tracking automatically reconstructs vehicle trajectories and matches them across drone views. Field experiments conducted at a freeway weaving section achieved over 91% accuracy in OD flow estimation compared with ground-truth data, demonstrating the framework’s effectiveness for scalable and high-resolution traffic monitoring. The second study proposes a day-specific spatial-temporal graph convolutional network (Day-STGCN) for real-time traffic forecasting. Building upon the classical STGCN architecture, the model integrates rolling historical time-dependent features with day-type categorization to capture both short-term variations and long-term recurring patterns. Using real-world sensor data, the model significantly improves prediction accuracy over baseline graph neural networks, providing robust day-aware forecasting that reflects realistic daily traffic dynamics. The third study investigates cooperative adaptive cruise control (CACC) and its effects on traffic string stability under varying market penetration rates of connected and automated vehicles (CAVs). Analytical derivations establish explicit stability conditions linking controller parameters, time headway, and control intervals in both homogeneous and heterogeneous traffic flows. The results reveal that appropriately designed CACC controllers enhance string stability even in mixed environments with different types of vehicles. Collectively, these studies present an integrated framework that advances ITS through data-driven sensing, learning, and control, offering theoretical and practical foundations for more intelligent, responsive, and stable transportation systems.

Computer vision has many applications that were implemented to the existing advance technology. Vehicle tracking and counting in video surveillance is one of them. Vehicle tracking allow the computer to detect and tracking the vehicles that capture through the video surveillance. Afterwards vehicle counting allow to the computer to count all of the vehicles on video. Background Subtraction is one of the method to perform object tracking and counting process for the system. This research implements Pixel Based Adaptive Segmenter Model for Background Subtraction method to perform object tracking and counting. This research used video capture of visual surveillance and vehicles as object to track and count. Result of this research is the system can track and count all vehicles on videos.

Intelligent Transport Systems consists of applications such as accident warning and avoidance systems. This is an area of interest to us in this research. The accident detection and alert system in the field of ITS (intelligent transportation system) are implemented and analyzed using many statistical methods for proving the accident detection point, GPS Coordinates points search, etc. In this paper, the analysis is performed on how accidents can be prevented using statistical analysis applied to the flow of vehicles in the construction area. This study helps to compare different statistical methods for vehicle detection in accident warning systems. Different statistical methods like Poisson’s probability distribution, Binomial distribution and Negative Binomial probability distributions were evaluated considering the count of vehicles and their distances from the area near or far from the prone area considered in the construction site. The aim of this paper is to evaluate and compare the statistical methods on the real-time datasets collected by hardware configuration of Arduino-Mega, Node-MCU and cloud service thingsspeak.com and shows that Poisson’s statistical distribution performs better than other methods in terms of probability of success. The different statistics like mean, standard deviation, and variance for different statistical methods are evaluated, out of these Poisson performs better.KeywordsAccident avoidanceIntelligent transportation system (ITS)Poisson’s distributionBinomial distributionNegative binomial distribution

In order to reduce traffic congestion and improve the efficiency of traffic light signals, intelligent traffic systems are being developed by researchers, and vehicle counting is one of the key techniques in the system. The traditional methods mostly focus on increasing the vehicle counting effectiveness without regard to the program execution efficiency. The practical value of these systems will be reduced if they cannot be operated in real-time on compact IoT device. Therefore, in this paper, we mainly focus on designing a real-time and robust system for the problem of counting specific-movement vehicles. The system is able to detect and track objects in the area of interest, then count those tracked trajectories using the movements. To improve performance of tracking multiple objects, a high recall detection method and an efficient feature matching strategy were proposed. Moreover, to minimize the wrong direction of movement prediction and improve the results of vehicle counting, a cosine similarity-based vehicle counting scheme is applied. Experiments are conducted on AI City 2021 Track-1 dataset. Our method is evaluated on both sides of efficiency and effectiveness.

Traffic congestion is a serious issue in developing nations. Light configurations are automatically modified by smart traffic light systems in response to traffic circumstances in real-time. For the system to adjust correctly, information on traffic density would be required. A vehicle counting system that can be used to determine the number of vehicles on busy highways with the aid of neural networks. YOLO (You Only Look Once), an object detection technique based on neural networks, is used by this system to identify cars. Simple Online and Real-time Tracking (SORT) algorithms are used to count and categories the vehicles in traffic recordings. We will determine the direction of vehicle movement after counting and attempt to apply the type of vehicle count, such as how many cars, how many bicycle and so on. The proposed neural network structure is better suited for real-time vehicle tracking because the computational complexity is reduced.

Intelligent transportation systems (ITS) are the integration of information and communications technologies with applications which are significant in traffic control and management. The increased number of on road vehicles in urban areas urges the need of development of automated methods for traffic management. Vehicle identification, classification and analysis enable the intelligent transportation systems to make decisions. In this paper, an automated method for video analysis for vehicle identification using a modified Region based Convolution Neural Network (RCNN) has been proposed. The traffic videos collected by CCTV cameras installed on the roads are analyzed for vehicle identification in a given frame. The pretrained google net is used to extract features. These features are used by the Region based Convolution Neural Network for vehicle identification. The vehicles are identified using probability score computed using intersection of objects (IoU). The identified vehicles are classified into ten different vehicle classes. The proposed network concatenates features from previous layers to reduce loss and consequently improve the vehicle identification accuracy. The vehicle identification method is further extended for vehicle counting and behavioral analysis. The vehicle counting information can be used for congestion control in smart cities. The behavioral analysis includes computation of speed of vehicles. The speed information is useful for traffic law enforcement in smart cities. The proposed method is applied on MIO-TCD vehicle dataset and EBVT video dataset. The results are calculated using three different metrics namely average accuracy, mean precision and mean recall. Obtained results are also compared with other state of the art methods. The results show significant improvement and thus the method can be effectively used for video analysis.

Traffic management systems require a large investment in developing the traffic management infrastructure. Most of this investment is focused on the overhead and underground cabling cost to provide connectivity to the central control station. This however, can be solved through the use of a low power wide area network (LPWAN). LPWAN’s require its nodes to be low power devices capable of decoding raw sensor data and send them through low bitrate and low bandwidth channels. This requires nodes in the traffic management system to be able to efficiently decode raw video data into usable traffic metrics such as vehicle count. This paper proposes a simple vehicle counting algorithm to minimize the processing power needed to provide accurate vehicle counts. This algorithm utilizes a foreground detector based on Gaussian mixed modelling and an IOU based tracker to perform the vehicle counting. The algorithm was tested on a video with 7 vehicles. The algorithm was able to detect 493 total detections and narrow it down to 9 unique tracklets. Overall, the system is promising as it was able to perform basic detection, segmentation and tracking at 132 fps with 78% accuracy.

This paper proposes an automatic vehicles counting, classification and tracking for identifying Traffic flows in an intersection which is a fundamental task for video surveillance in urban traffic management. In intelligent transportation systems (ITS), especially in field of urban traffic management, intersections monitoring is one of the critical and challenging tasks. Where, objects have different characteristics such as varying velocities, stop and go. Therefore it is necessary to use a hybrid scheme of adaptive background mixture model to learn background model faster and more accurately, instead of using single rate of adaptation, which is not adequate. The main focus of this paper is to analyze activities at intersection in number of traffic zones for counting and classifying vehicles and then tracking to extract traffic flows which assists in regulating traffic lights for using in smart cameras. Traffic zones definition in intersection video, based on trajectories clustering; greatly reduce the time and volume of computations. Smart camera's fundamental purpose is to efficiently reduce the transmission rate and also analyze an intersection scene and report statics and information of interest which is not an image. Ground-truth experiments with urban traffic sequences show that our proposed algorithm is very promising relative to results using other techniques.

In developing countries like India, limited Intelligent Transportation Systems (ITS) infrastructure is available and it is difficult to generate real time traffic information for city wide or larger regions. On the other hand, cellular network is widely deployed in India covering major part of the road network. However, cellular network based positioning data have large location error (250-500 meters) and is not directly useful for edge level travel time or speed estimation. Also, penetration of GPS enabled vehicles/phones is very low in India to generate useful real time traffic information. In this simulation based study we use the map matching algorithm, proposed in [1], that operates on a series of position points having large location error to get accurate vehicle count for every edge, with time lag of about 10 minutes. Temporal extrapolation using exponential moving average is used to overcome the time lag. The vehicle count is then used to estimate edgewise vehicle flow and occupancy. Simulation results show the mean estimation error of less than 10%. Then, the edge level congestion estimation is done using volume-capacity ratio (V/C ratio) and mean error of less than 10% is observed. These results are encouraging as they show feasibility of using widely deployed cellular network for real time congestion level estimation. For edge level speed estimation, a subset of edges are chosen for infrastructure deployment using historical data of congestion levels observed on various edges in a road network. The edges with ITS infrastructure are used to learn the occupancy-speed relationship which is then spatially extrapolated to infrastructure-less edges using GPS probe data to enable travel speed estimation on all the edges in the road network. The simulation results show that speed estimation with ninety percentile error of less than 15% is achievable.

Traffic density data plays an important role in decision making by the Intelligent Transportation System (ITS). This system uses this data in the process of adaptive traffic management. The inaccuracy of the data provided into the ITS system can result in errors in decision making. This study utilizes digital image engineering technology in the detection of four-wheeled vehicles in traffic traffic for the purpose of acquiring traffic density data. In this study, we propose a multi-ROI (pixel area methodRegion of Interest). This multi-ROI proposal is to be put forward to improve reading accuracy compared to just one ROI. With the use of this multi-ROI, the information obtained from the overall ROI can strengthen the accuracy of the data of vehicles passing in a lane. Our experimental results show that the use of multi-ROI with a certain amount of ROI can produce an accuracy rate of up to 88.66% compared to single-ROI which has an accuracy rate of 84.65%.

The rapid advancement in the field of deep learning and high performance computing has highly augmented the scope of video-based vehicle counting system. In this paper, the authors deploy several state-of-the-art object detection and tracking algorithms to detect and track different classes of vehicles in their regions of interest (ROI). The goal of correctly detecting and tracking vehicles’ in their ROI is to obtain an accurate vehicle count. Multiple combinations of object detection models coupled with different tracking systems are applied to access the best vehicle counting framework. The models’ addresses challenges associated to different weather conditions, occlusion and low-light settings and efficiently extracts vehicle information and trajectories through its computationally rich training and feedback cycles. The automatic vehicle counts resulting from all the model combinations are validated and compared against the manually counted ground truths of over 9 h’ traffic video data obtained from the Louisiana Department of Transportation and Development. Experimental results demonstrate that the combination of CenterNet and Deep SORT, and YOLOv4 and Deep SORT produced the best overall counting percentage for all vehicles.

Automatic data labeling by neural networks for the counting of objects in videos

Vehicular Adhoc NETworks (VANETs) is a new and emerging technology for wireless communications that has attracted considerable attention in the last years from the academic, scientific, industrial and governmental communities, due to the improvements and the new features that it brings to the Intelligent Transportation Systems (ITSs). In this paper, we present an algorithm that uses VANET technology to determine the total number of vehicles that are stopped at a traffic light at an intersection. To facilitate an efficient counting, we divide the road into fixed-size road regions and through a leader election mechanism, we designate one vehicle in each region as the “Region Leader” that is in charge of computing and propagating the total number of nodes in its region. Additionally, the region leader will act as a router to retransmit counting information from other region leaders that are further away, so it can eventually reach the central processing point where it is processed. We have carried out extensive experiments in various scenarios to validate and analyze the behavior and precision of our new proposed algorithm. Also, we compare our algorithm with another strategy proposed by a research team led by Alok Rajan. Our simulations were done using Veins (Vehicle in Network Simulation), an open-source simulation framework that ties together two simulators: OMNeT++ for the network simulation and SUMO for the microscopic road traffic simulation. Both simulators run in parallel and communicate with each other through a TCP connection using a protocol called TraCI (Traffic Control Interface). Our simulation results show that the algorithm performs an effective counting of vehicles, with a reduced response time, and a small total number of control messages sent by the vehicles to accomplish the counting task, under different conditions of vehicular traffic loads.