No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Real-time image processing approach to measure traffic queue parameters
Abstract
The real-time measurement of various traffic parameters including
queue parameters is required in many traffic situations such as accident
and congestion monitoring and adjusting the timings of the traffic
lights. In case of the queue detection, at least two algorithms have
been proposed by previous researchers. Those algorithms are used for
queue detection and are unable to measure queue parameters. The authors
propose a method based on applying the combination of noise insensitive
and simple algorithms on a number of sub-profiles (a one-pixel-wide
key-region) along the road. The proposed queue detection algorithm
consists of motion detection and vehicle detection operations, both
based on extracting edges of the scene, to reduce the effects of
variation of lighting conditions. To reduce the computation time, the
motion detection operation continuously operates on all the
sub-profiles, but the vehicle detection is only applied to the tail of
the queue. The proposed algorithms have been implemented on an
80386-based microcomputer system and the whole system works in real-time
... Traffic incidents comprise excessive vehicle speed and length, illegal conversions and parking, driving against the oncoming traffic, accident detection and location, queue detection and estimation of its length, and lane occupancy and flow that typify traffic congestion. Distinct approaches to extracting useful traffic information from image sequences have been employed such as monochromatic intensity and background differencing for motion detection and segmentation (Rosin and Ellis, 1995;Fathy andSiyal, 1995, 1998), monochromatic edge counting (Fathy andSiyal, 1995, 1998) and chroma (Prati et al., 2001;Horprasert et al., 1999) for vehicle detection, and corner-and region-based tracking (Coifman et al., 1998). Multiple model hypothesis and likelihood ratio tests to yield estimates of traffic parameters, and algorithms for traffic congestion detection and location to provide decision support are described and evaluated in (Kastrinaki et al., 2003;Weil et al., 1998;Willsky et al., 1980). ...
... Traffic incidents comprise excessive vehicle speed and length, illegal conversions and parking, driving against the oncoming traffic, accident detection and location, queue detection and estimation of its length, and lane occupancy and flow that typify traffic congestion. Distinct approaches to extracting useful traffic information from image sequences have been employed such as monochromatic intensity and background differencing for motion detection and segmentation (Rosin and Ellis, 1995;Fathy andSiyal, 1995, 1998), monochromatic edge counting (Fathy andSiyal, 1995, 1998) and chroma (Prati et al., 2001;Horprasert et al., 1999) for vehicle detection, and corner-and region-based tracking (Coifman et al., 1998). Multiple model hypothesis and likelihood ratio tests to yield estimates of traffic parameters, and algorithms for traffic congestion detection and location to provide decision support are described and evaluated in (Kastrinaki et al., 2003;Weil et al., 1998;Willsky et al., 1980). ...
... Incident detection often demands accurate segmentation of a vehicle from the lane background regardless of shadows and varying illumination. This paper probes further into and extends to edge magnitude and chroma the results in Fonseca and Waldmann (2004) on nonparametric procedures for automatic edge-count threshold adaptation in a monochromatic vehicle detection algorithm that evolved from previous work by Fathy andSiyal (1995, 1998). This paper proposes the use of intensity edges and chromatic distortion to attain robust performance and reduce the need for operator interventions. ...
Computer vision algorithms for traffic incident detection often demand accurate segmentation of a vehicle regardless of shadows and varying illumination. This paper proposes the use of intensity and chroma information to attain robust perform-ance, and reduce the need for operator interventions. The major contribution is the use of a few ad hoc rules to detect a vehicle based on learned lane background intensity levels, detected edge pixels, and chromatic distortion. Two algorithms for vehicle de-tection are proposed and investigated, each one composed of two stages. The first stage is training, when progressive background learning and threshold adaptation occur within a region of interest (ROI). The second stage is the detection of a vehicle within the ROI and the computation of corresponding traffic parameters. Tests with an actual image sequence rich in moving shadows indi-cate that the processing of chroma and monochromatic data in one of the proposed algorithms reduces false alarms while main-taining an acceptable detection rate in comparison with a monochromatic, edge-count algorithm cited in the literature and em-ployed here as the reference algorithm. The other proposed algorithm is monochromatic only, based on edge magnitude, and im-proves detection rate and false alarm with respect to the reference algorithm at the expense of a more complex threshold learning stage but without imposing an additional computational burden during the detection stage.
... Vision-based queue analysis can be performed by two major groups of tracking and non-tracking methods. Nontracking methods determine the existence of vehicles based * 1 M. S. Shirazi [6], spatial edges [8], FFT [8], and image gradients [21] are the important features used to detect stopped vehicles in the literature. As an example of other features, the entropy method is proposed [20] to detect stopped vehicles and Harris corner features [1] are useful in detecting stopped vehicles when they build a queue. ...
... Vision-based queue analysis can be performed by two major groups of tracking and non-tracking methods. Nontracking methods determine the existence of vehicles based * 1 M. S. Shirazi [6], spatial edges [8], FFT [8], and image gradients [21] are the important features used to detect stopped vehicles in the literature. As an example of other features, the entropy method is proposed [20] to detect stopped vehicles and Harris corner features [1] are useful in detecting stopped vehicles when they build a queue. ...
This paper presents a tracking method for vision-based queue analysis at junctions including queue length and waiting time estimation of vehicles. The tracking method works based on improving the optical flow method to track small and low quality vehicles with overlong waiting time. The improvement process is performed to keep track of overlong stopped vehicles with high level of robustness against occlusion by crossing pedestrians. The results of experiments are presented by estimating queue length, waiting time distribution and number of waiting vehicles for the highly cluttered video of a Las Vegas junction. The accuracy of the system is evaluated by comparing the queue analysis results with the ground truth.
... Video cameras are not only cheaper, but simpler to install and maintain. Computer vision is widely applied in transportation systems, such as traffic congestion detection [2,3], queue length measurement at traffic lights [4,5,6,7], lane occupancy estimation [8], vehicle classification [9,10], and trajectory learning and prediction. ...
... Fathy [4] combines the three methods to measure the queue and delay length. The time difference and background difference methods detect motion in the scene by identifying a deviation in the intensity value of the same pixel in two different captures. ...
This study aims to estimate the traffic load at street intersections obtaining the circulating vehicle number through image processing and pattern recognition. The algorithm detects moving objects in a street view by using level lines and generates a new feature space called movement feature space (MFS). The MFS generates primitives as segments and corners to match vehicle model generating hypotheses. The MFS is also grouped in a histogram configuration called histograms of oriented level lines (HO2 L). This work uses HO2 L features to validate vehicle hypotheses comparing the performance of different classifiers: linear support vector machine (SVM), non-linear SVM, neural networks and boosting. On average, successful detection rate is of 86% with 10-1 false positives per image for highly occluded images.
... However, this technique has problems of accurately updating the background frame and automatically selecting a suitable threshold value. Edge detection based segmentation of traffic scene has the advantage of being less sensitive to variation of ambient lighting and shadows [6]. However, combined background differencing and edge detection has the advantage of eliminating stationary vehicles, shadows and the road markings and is less sensitive to variations of lighting [2]. ...
... This vehicle detection technique has better performance than the background differencing operations as colours of vehicles and the ambient lighting changes in traffic scenes are less sensitive to edge detection. Following the application of vehicle detection operation, the number of pixels having greater value than the threshold is used to recognise a vehicle [6]. ...
Various researchers for a number of years have investigated traffic data collection and analysis. Many techniques have been proposed to speed up the operations and some intelligent approaches have been developed to compensate the effects of lighting, shadows and occlusions. To achieve real-time processing, it is necessary to reduce the amount of data to be processed, so the first question is to determine the suitable location for the key regions or windows. Previously, researchers have either used full frame processing approach, which requires more computing power and thus is not practical for real-time applications, or have used window based techniques, but defining of windows was done manually, which is not practical in real-world traffic applications. In this paper, an automatic approach is described to measure the size and location of windows required for this purpose. The results demonstrate that this method provides better results than the fixed window size.
... Based on the current traffic circumstances, this sort of traffic management system cannot prolong the current green light time length and cannot minimize the vehicle's waiting time at the red light. A fuzzy logic-based traffic light controller may be utilized for optimal regulation of traffic volumes, such as oversaturation or exceptional load circumstances [20,21,22,23,24,25,26,27,28]. The paper proposes the deployment of a developed fuzzy logic controller system for regulating traffic signals at numerous intersection models in the city. ...
Traffic congestion difficulties have resulted in low productivity, significant air pollution, and energy losses in Owerri Metropolis, Nigeria. The paper looked at the design of a smart fuzzy logic traffic light management module, the development of a traffic control program using an Arduino microcontroller system, and the validation of the developed program's functionality using a Proteus circuit model to confirm the efficiency of fuzzy signal control. The traffic light environment of a fuzzy logic controller is simulated using Matlab software, and isolated traffic of multiple junctions is simulated using the Sumo Urban Mobility Simulation (SUMO) environment. The performance of the traditional fixed-time controller vs the fuzzy logic traffic light controller is examined. To allocate pedestrian crossing the right of way, traffic light systems function in tandem with pedestrian displays. The simulation results indicated that the overall durations for the fixed traffic light controller and suggested smart fuzzy logic traffic light controller simulations are 1,426 seconds and 1,328 seconds, respectively. The results also showed that the suggested smart fuzzy logic traffic light controller recorded significant waiting and movement times, indicating its standard appropriateness for Owerri city's various crossings.
... Based on the current traffic circumstances, this sort of traffic management system cannot prolong the current green light time length and cannot minimize the vehicle's waiting time at the red light. A fuzzy logic-based traffic light controller may be utilized for optimal regulation of traffic volumes, such as oversaturation or exceptional load circumstances [20,21,22,23,24,25,26,27,28]. The paper proposes the deployment of a developed fuzzy logic controller system for regulating traffic signals at numerous intersection models in the city. ...
Traffic congestion difficulties have resulted in low productivity, significant air pollution, and energy losses in Owerri Metropolis, Nigeria. The paper looked at the design of a smart fuzzy logic traffic light management module, the development of a traffic control program using an Arduino microcontroller system, and the validation of the developed program's functionality using a Proteus circuit model to confirm the efficiency of fuzzy signal control. The traffic light environment of a fuzzy logic controller is simulated using Matlab software, and isolated traffic of multiple junctions is simulated using the Sumo Urban Mobility Simulation (SUMO) environment. The performance of the traditional fixed-time controller vs the fuzzy logic traffic light controller is examined. To allocate pedestrian crossing the right of way, traffic light systems function in tandem with pedestrian displays. The simulation results indicated that the overall durations for the fixed traffic light controller and suggested smart fuzzy logic traffic light controller simulations are 1,426 seconds and 1,328 seconds, respectively. The results also showed that the suggested smart fuzzy logic traffic light controller recorded significant waiting and movement times, indicating its standard appropriateness for Owerri city's various crossings.
... The efficiency of the traffic light in the queue model however, was affected by the occurrence of unexpected events such as the break-down of a vehicle or road traffic accidents thereby causing disruption to the flow of vehicles. Among those techniques based on the queue model was a queue detection algorithm proposed by [8]. The algorithm consisted of motion detection and vehicle detection operations, both of which were based on extracting the edges of the scene to reduce the effects of variations in lighting conditions. ...
... Various systems were implemented to measure this parameter. However, these systems are based on traditional image processing algorithms for detecting vehicles such as edge detection (Slimani et al., 2018) (Al Okaishi et al., 2019a) (Fathy and Siyal, 1995), and corner detection (Albiol et al., 2011a) (Chintalacheruvu and Muthukumar, 2012) (Albiol et al., 2011b). Due to the utilization of these algorithms, the performance of these systems can be affected by several issues such as weather conditions, shadows, the marks on the roads, and etc. ...
Vehicular queue length measurement is an important parameter to detect the traffic congestion, which is resulted from several issues such as traffic lights, accidents, and poor roads infrastructures. In this paper, a system in real-time is proposed to detect and measure the vehicular queue length at intersections. The proposed system consists of two main steps: the first step is the detection of queue by using frames differencing method to detect the motion in the target areas. If there is no a motion, then the second step is implemented to detect the vehicles in these areas by using Single Shot Multibox Detector (SSD) algorithm. If there are vehicles, that means the queue exists and the measurement process begins. Some modifications are applied on SSD algorithm to fit with in our system and to improve the accuracy of the vehicle detection process. The system is applied on videos obtained by stationary cameras. The experiments demonstrate that this system is able to accurately detect and measure the vehicular queue length.
... Specifically, we are interesting in the local approach that describe the principle of a queue processing technique, where the cars leave the queue in the order they arrive, or waiting one's turn at a traffic control signal. Fathy and Siyal [20] proposed a queue detection algorithm consisting of motion detection and vehicle detection operations, both based on extracting edges of the scene, to reduce the effects of variation of lighting conditions. To reduce the computation time, the motion detection operation continuously operates on all the sub-profiles, but the vehicle detection is only applied to the tail of the queue. ...
The traffic signal control plays a fundamental role to improve the efficiency and efficacy of traffic flows in traffic networks. This paper is the first work in which we consider a mathematically rigorous study of the continuous-time, discrete state, multi-traffic signal control problem using a non-cooperative game theory approach. The solution of the problem is circumscribed to an ergodic, controllable, discrete state, continuous-time Markov game computed under the expected average cost criterion. This paper provides several main contributions. First, we present a general continuous-time queue model, which is employed as the fundamental scheme of a computationally tractable game theory approach for the signal control continuous-time Markov game. This model is transformed into a discrete state Poisson process where the vehicles leave the queue in the order they arrive. Second, in this problem, each signal controller (player) aims at finding green time that minimizes its signal and queuing delay. Then, a conflict appears when each signal controller tries to minimize its queue. We study the problem of computing a Nash equilibrium for this game. Our third contribution employs a proximal/gradient method for computing the Nash equilibrium point of the game. By introducing new restrictions over the signal controller and adding a restriction for continuous-time Markov chains, we obtain the set of average optimal policies, which is one of the main results of this paper. Hence, our final contribution shows, in simulation, the usefulness of the proposed method with an application example.
... Different techniques exist to determine lengths of the queue in each lane on street width and the number of vehicles that are expected at a given time of day. Fathy el al [7] proposed a queue detection algorithm based on motion detection and vehicle detection algorithm. The proposed hybrid system has two more components namely the Accident Detection System and Action System. ...
The development of traffic signal control systems has become one of the most important topics in this era. Traffic light controllers need to be improved continuously to solve the traffic problems. This paper discussed the proposed hybrid system and demonstrated how the system works from the beginning of the first flag “decrease of cross ratio” until the end of the action system. The proposed system was divided into three main parts: The proposed algorithm (Dynamic Webster with dynamic Cycle Time), Accident Detection System using fuzzy logic theory and Action System depending on Detection System. The focus of this paper is to discuss the accident detection system of the proposed hybrid system, which depended on fuzzy logic and its components. This paper also presented the results of FuzzyTech Software with different scenarios plotting the inputs outputs and the showcases the 3D plot for each one of them for detecting the accident. In addition, it presented results to measure the False Alarm Rate the Accident Detection Rate using FuzzyTech program.
... Different techniques exist to determine lengths of the queue in each lane on street width and the number of vehicles that are expected at a given time of day. Fathy el al proposed a queue detection algorithm based on motion detection and vehicle detection algorithm [7]. The proposed hybrid system has two more components namely the Accident Detection System and Action System. ...
Development of Traffic signal control systems has become one of the most important topics in this era. Traffic light controllers need to be developed continuously to solve the traffic problems. This paper discussed the proposed Hybrid system and demonstrated how the system works from the beginning of the first flag “decrease of cross ratio” until the end of the action system. The proposed system was divided into three main parts: The proposed algorithm (Dynamic Webster with dynamic Cycle Time), Accident Detection System using fuzzy logic theory and Action System depending on Detection System. The focus of this paper is to discuss the Accident Detection System of the proposed Hybrid system, which depended on fuzzy logic and its components. This paper also presented the results with scenario plotting the inputs outputs and the showcases the 3D plot for the scenario to detect the accident using FuzzyTech program.
... Different techniques exist to determine lengths of the queue in each lane on street width and the number of vehicles that are expected at a given time of day. Fathy el al proposed a queue detection algorithm based on motion detection and vehicle detection algorithm [7]. The proposed hybrid system has two more components namely the Accident Detection System and Action System. ...
Development of Traffic signal control systems has become one of the most important topics in this era. Traffic light controllers need to be developed continuously to solve the traffic problems. This paper discussed the proposed Hybrid system and demonstrated how the system works from the beginning of the first flag "decrease of cross ratio" until the end of the action system. The proposed system was divided into three main parts: The proposed algorithm (Dynamic Webster with dynamic Cycle Time), Accident Detection System using fuzzy logic theory and Action System depending on Detection System. The focus of this paper is to discuss the Accident Detection System of the proposed Hybrid system, which depended on fuzzy logic and its components. This paper also presented results using iTraffic Simulation to measure the False Alarm Rate the Accident Detection Rate.
... Different techniques exist to determine lengths of the queue in each lane on street width and the number of vehicles that are expected at a given time of day. Fathy el al proposed a queue detection algorithm based on motion detection and vehicle detection algorithm [7]. ...
Development of Traffic signal control systems has become one of the most important topics in this era. Traffic light controllers need to be developed continuously to solve the traffic problems. This paper discussed the proposed accident detection system and demonstrated how the system works from the beginning of the first flag "decrease of cross ratio" and through the member functions with the status of the accident. The focus of this paper is to discuss the Accident Detection System of the proposed dynamic fuzzy control system, which depended on fuzzy logic and its components.
... The efficiency of the traffic light in the queue model however, was affected by the occurrence of unexpected events such as the break-down of a vehicle or road traffic accidents thereby causing disruption to the flow of vehicles. Among those techniques based on the queue model was a queue detection algorithm proposed by [8]. The algorithm consisted of motion detection and vehicle detection operations, both of which were based on extracting the edges of the scene to reduce the effects of variations in lighting conditions. ...
This paper described our research experiences of building an intelligent system to monitor and control road traffic in a Nigerian city. A hybrid methodology obtained by the crossing of the Structured Systems Analysis and Design Methodology (SSADM) and the Fuzzy-Logic based Design Methodology was deployed to develop and implement the system. Problems were identified with the current traffic control system at the '+' junctions and this necessitated the design and implementation of a new system to solve the problems. The resulting fuzzy logic-based system for traffic control was simulated and tested using a popular intersection in a Nigerian city; notorious for severe traffic logjam. The new system eliminated some of the problems identified in the current traffic monitoring and control systems.
... SPITS measures a queue length in meters, but cannot provide the number of vehicles in a queue. Fathy and Siyal (1995, 1998) also developed image processing systems to measure volume, speed, vehicle length, and queue length. The profiles used to detect queue length were divided into subprofiles, each with approximately the same length per vehicle, thereby making it possible to estimate the number of vehicles in the queue. ...
Safety and quality of travel on arterial networks tie closely to the performance of signalized intersections. Measures commonly used for intersection performance evaluations are control delay, queue length, and cycle failure. However, these variables are not directly available from typical configurations of traffic sensors designed for intersection signal control. Collecting vehicle control delay data manually for intersection performance measurement has been a task too time-consuming and labor-intensive to be practical. Video image processors (VIPs) have been increasingly deployed for intersection signal control in recent years. This study aims to use the extra detection capabilities of VIPs for performance monitoring at signalized intersections. Most VIPs can support up to 24 virtual loops, but normally less than half of the virtual loops are used. By properly configuring the spare virtual loops and analyzing the loop measurements, intersection performance can be monitored in real time. In this research, we propose an approach for measuring queue length and vehicle control delay at signalized intersections based on traffic count data collected with traffic sensors. This algorithm has been implemented in a computerized system called In-PerforM. The In-PerforM system was evaluated by both field tests and simulation experiments. Although the VIPs’ counting errors do affect the accuracy of field test results, we still received encouraging results on queue lengths and control delay measurements in both the field tests and simulation experiments. This demonstrates that the In-PerforM system, and therefore the proposed algorithm, has the potential to be a cost-effective approach for performance measurement at signalized intersections.
... In [11] and [12] authors used the background estimation technique with an efficiency rate over 90% in both cases. On the other hand, in [13] a morphological edge detector (SMED) was developed, which presents higher insensitivity to illumination changes than the background estimation, obtaining an accuracy of 95%. Figure 2 shows the physical diagram of the proposed solution. A computer is placed at each intersection, which is in charge of acquiring images from a network of cameras. ...
RESUMEN: Este artículo presenta el desarrollo de un controlador de tráfico difuso capaz de gestionar de manera autónoma, centralizada y eficiente, el flujo vehicular en un grupo de intersecciones. El sistema emplea un algoritmo de visión artificial que le permite detectar el
número de autos presentes en imágenes capturadas por un conjunto de cámaras estratégicamente ubicadas en cada intersección. Usando esta información, el sistema selecciona la secuencia de acciones que optimicen el flujo vehicular dentro de la zona de control, en un escenario simulado. Los resultados obtenidos muestran que el sistema disminuye en un 20% los tiempos de retraso para cada vehículo y que además es capaz de adaptarse rápida y eficientemente a los cambios de flujo.
... P. Briquet and J. Versavel [9][10] put forward a method which combines image frames differing and hardware [11] proposed a method which combines image frames differing with a macroscopic analysis of the traffic flow to detect whether the vehicle is in motion or at rest, and then determine the vehicle queuing situation. Y. M. Fathy and M. Siyal [12] put forward a detection algorithm combining the queuing length, queuing status, possession cycle and occupancy. This method does image frames differing in closed small areas on a onepixel-wide outline. ...
In order to provide better traffic planning, monitoring for road traffic is very necessary. In this paper, interframe difference is mainly studied for extracting moving vehicles and clustering analysis is used to get the target number. DBSCAN clustering analysis is the key to deal with the problem to get the number of the targets. In order to achieve better clustering effect, this paper uses the median filter and mathematical morphology filtering. According to the target state change, traffic statistics has been done. Meanwhile, parameters such as searching area, clusters threshold and frame differing threshold are made adaptive in this paper. In order to improve real-time performance, sampling is adopted to each frame. By analyzing a streaming video, the algorithm for traffic statistics can achieve good results.
... At present, the limitations when the video-based queue length detection technology put into practice include: 1), the threshold setting is difficult; 2), rate of false alarm is high; 3), the precision of acquired data is low and influenced by external environmental interference; 4), large amount of data causes many transmission and processing problems; 5), commonality of data and algorithms is not high, and it is not easy for portability and adaptability. Many scholars have dedicated to research on these issues, such as Hoose N [5], Rouke A and Bell M G H [6], Fathy M and Siyal M Y [7] [8], Li and Zhang [9] and so on[10][12], and they have proposed a lot of methods and algorithm frameworks to improve these limitations. These studies promote the development and progress of embedded vehicle queue length detection. ...
The vehicle queue length detection, specially based on video, is an important and practical research area of Intelligent Transportation Systems (ITS). It has promising application prospect in such areas as urban traffic control, highway toll system etc. In this paper, by analyzing the video sequence obtained from the fixed-camera, the vehicle movements of each road lane, such as left turn, go straight, turn right, are detected by using telescopic virtual coil, and then the vehicle motion detection and vehicle presence detection are used to calculate the vehicle queue length. On this basis, the vehicle queue detection algorithms are improved, and their threshold values are modified in the paper correspondently. And, the DM642 platform is designed to implement the real-time detection of vehicle queue length on the crossroads. Application experiments prove that the solution can detect the vehicle queue length with high precision, and the detection results meet the accuracy and stability requirements of practical application.
... The most significant disadvantage of these sensors lies in the fact that they can survey only a limited region of traffic path and cannot give an image of the whole path. A ma major idea in this paper, which was initiated by some researchers [6,7], is to analyze a wider view of the path and evaluate a whole description of traffic status. In this paper we have concentrated on describing image-processing algorithms together with the results for qualitative road traffic analysis. ...
Real-time qualitative road traffic data analysis is the cornerstone for any modern transport system. So far, most of the analysis is done manually and the use of image processing techniques for qualitative analysis is at its early stage. In this paper we concentrate on qualitative analysis of a traffic scene, which is describing the traffic scene similar to the description of a traffic expert when viewing the scene. The results demonstrate that the system can be used in real-world traffic situation and the results are reported on-line in a graphical form.
... In [12] and [13] authors used the background estimation technique with an efficiency rate over 90% in both cases. On the other hand, in [14] a morphological edge detector (SMED) was developed, which presents more insensitiveness to illumination changes than the background estimation, obtaining an accuracy of 95%. Figure 2 shows a physical diagram of the solution. At each intersection a computer is placed, this is in charge of acquiring images from a network of cameras. ...
This paper presents a fuzzy traffic controller that in an autonomous, centralized and optimal way, manages traffic flow in a group of intersections. The system obtains information from a network of cameras and through machine vision algorithms can detect the number of vehicles in each of the roads. Using this information, the fuzzy system selects the sequence of phases that optimize traffic flow globally. To evaluate the performance of the controller, a scenario was developed where it was possible to simulate through artificially created videos two adjacent intersections. System performance was compared versus fixed time controllers as they are currently the most used in the city of Bogota. As a control variable it was used the average waiting time of each vehicle. The results show that the system performance increases by about 20% over situations with heavy traffic conditions and that the controller is able to adapt smoothly to different flow changes.
... In this paper we extend STEMS with capabilities to support the dynamic aspect of the traffic environment by adapting to the continuously changing traffic conditions, as captured by various sensors and surveillance technologies. The ubiquitous deployment of sensors in the transportation network, along expressways131415 and at intersections of local streets [4, 5, 12], is only one of the motivating factors for our work. The second important factor is the increased accessibility of the collected real-time and historical traffic data over the web. ...
Contemporary terrorists have made public transportation a new theater of operations. Specifically, attacks on the urban transportation system can cause great disruption and alarm, which are the traditional goals of terrorism. The unexpected and stochastic character of terrorist attacks poses unique challenges to those responsible for security. Since the threat of terrorism is obscure and security measures are costly, it is hard to justify the expenditures before an attack. Security against terrorism therefore tends to be reactive. In this paper we propose new ITS technologies to enhance the surface transportation aspects of homeland security, by providing more efficient and safer evacuation for general public in case of terrorist attacks or other human caused disasters. In particular, we extend our existing work on developing a smart traffic evacuation management system (STEMS), by enhancing it with capabilities to adapt to the dynamics of the traffic environment, by leveraging real-time information obtained from sensors or other surveillance technologies. STEMS handles the unexpected aspect of terrorist attacks or other unpredictable disasters, by generating evacuation plans dynamically, when given an incident location and scope. To handle traffic dynamics, STEMS will continue to revise the initially generated plan, during the evacuation operation, to keep it consistent with the continuously changing traffic conditions. The advantages of this revision process are two-fold: first, it ensures that traffic will not be directed towards congested areas; and secondly, it decreases congestion by spreading the traffic from currently congested segments to alternative routes. Our simulation studies show that employing real-time information greatly improves STEMS performance and therefore, evacuation efficiency.
The world has been overtaken by artificial intelligence (AI), machine learning, and deep learning, which are affecting every aspect of our lives, including our need for companionship and the growth of smart cities and communities. Though tremendous progress has been made in fields like healthcare and education, transportation. Transportation is a significant barrier that has a huge financial and human cost on a worldwide scale. Traditional methods of assessing traffic on roads, which rely on inductive loops, don't give much information and can't control traffic in real-time. This study describes how the transportation sector could undergo a deep learning revolution. Deep learning techniques, including image and video analysis, have the potential to revolutionize road traffic monitoring by enabling computers to recognize traffic congestion, enhance road safety, and address various other concerns. The integration of deep learning technologies promises a new era of data-driven solutions for transportation-related issues, which could ultimately result in safer, more effective, and economically sustainable societies in the face of over a million annual fatalities and trillions of dollars in damages. In order to address these problems, this paper shows how well YOLOv3 detects automobiles and how well Darknet53 tracks these vehicles while they are on the road.
6 Şubat 2023 tarihinde Kahramanmaraş Pazarcık ilçesinde TS 04:17’de meydana gelen 7.7M w
büyüklüğündeki deprem ve aynı gün içerisinde TS 13:24’te Elbistan merkezli 7.6M w büyüklüğündeki
deprem sonucu 50.096 vatandaşımız vefat etmiş (AFAD, 2023) birçok yapı yıkılmış, ağır hasar almış ve
kullanılamaz duruma gelmiştir. 9 saat arayla olan ve 11 şehri etkileyen bu depremler sonucunda yıkılan
ve ağır hasarlı teknik raporu verilen yapıların değerlendirilmesi; meslek etiği kavramını, malzeme
kalitesini ve inşaat işlerinde uygulama hatalarını gündeme getirmiştir. Çevre, Şehircilik ve İklim
Değişikliği Bakanlığı Yapı İşleri Genel Müdürlüğü’nün son verilerine göre 197 bin 825 bina az veya orta
hasarlı, toplam yıkılan bina sayısı 18 bin 200 ve acil yıkılacak ağır hasarlı ve yıkılmış işyeri sayısı ise 61
bin 890 olarak kayılara geçmiştir. Bu kapsamda çalışmada yıkılan ve ağır hasar alan yapıların
bulundukları zemin geoteknik özellikleri, zemine uygun temel seçimi, taşıyıcı sistem düzensizlikleri,
inşaat malzeme kaliteleri ve şantiye uygulaması sırasında gerçekleşen hataların değerlendirilmesi
yapılmıştır.
2) 6: 581-583] © 2016 IJSRSET | Volume 2 | Issue 6 | Print ABSTRACT This paper focuses on improving vehicular traffic flow using neural network controller system. Such real time simulation software was currently used as a tool for optimizing the design of vehicular controls in traffic related matters. In this paper, field data were collected from Digital Security Company Enugu which included measurement from the average waiting time for the red light duration and different green light durations. The testbed environment is made up of a junction, ABCD. It was observed that the morning hour otherwise known as busy hour, more vehicles enters into the testbed environment through lane A and B while less vehicle queue in lane C and D respectively. With the introduction of neural network in the design, this will help in decongesting the vehicular problem both in the busy hour and less busy hour. This characterisation of testbed environment were neural network is used is not so with junctions where traditional traffic flow control is used, a lane with no vehicle on it is still has time allotted to it while those with long queues of vehicles are asked to stop and wait until they are asked to move.
Computer and Machine Vision: Theory, Algorithms, Practicalities (previously entitled Machine Vision) clearly and systematically presents the basic methodology of computer and machine vision, covering the essential elements of the theory while emphasizing algorithmic and practical design constraints. This fully revised fourth edition has brought in more of the concepts and applications of computer vision, making it a very comprehensive and up-to-date tutorial text suitable for graduate students, researchers and R&D engineers working in this vibrant subject. Key features include: Practical examples and case studies give the 'ins and outs' of developing real-world vision systems, giving engineers the realities of implementing the principles in practice. New chapters containing case studies on surveillance and driver assistance systems give practical methods on these cutting-edge applications in computer vision. Necessary mathematics and essential theory are made approachable by careful explanations and well-illustrated examples. Updated content and new sections cover topics such as human iris location, image stitching, line detection using RANSAC, performance measures, and hyperspectral imaging. The 'recent developments' section now included in each chapter will be useful in bringing students and practitioners up to date with the subject. Roy Davies is Emeritus Professor of Machine Vision at Royal Holloway, University of London. He has worked on many aspects of vision, from feature detection to robust, real-time implementations of practical vision tasks. His interests include automated visual inspection, surveillance, vehicle guidance and crime detection. He has published more than 200 papers, and three books - Machine Vision: Theory, Algorithms, Practicalities (1990), Electronics, Noise and Signal Recovery (1993), and Image Processing for the Food Industry (2000); the first of these has been widely used internationally for more than 20 years, and is now out in this much enhanced fourth edition. Roy holds a DSc at the University of London, and has been awarded Distinguished Fellow of the British Machine Vision Association, and Fellow of the International Association of Pattern Recognition. Mathematics and essential theory are made approachable by careful explanations and well-illustrated examples. Updated content and new sections cover topics such as human iris location, image stitching, line detection using RANSAC, performance measures, and hyperspectral imaging. The 'recent developments' section now included in each chapter will be useful in bringing students and practitioners up to date with the subject.
In this chapter, intersection analysis including capacity, delay, and safety is presented using computer vision techniques. An intersection appropriate vision-based tracking system is presented, which aims to provide long-term tracks of road users provide classification (i.e., vehicle and pedestrian) and handle the partial occlusion problem. Road trajectories are further investigated and modeled to provide road user count, vehicle queue length, and safety analysis including accidents and conflicts. Since accidents are infrequent events, surrogate safety measurements were leveraged to provide conflict severity measures at intersection facilities. Finally, technology-enhanced safety for all participants, including vehicles, drivers, and pedestrians, through communication and sharing of dynamic profiles between infrastructure and cooperative vehicles is highlighted.
The traffic signal control setting is the most important factor that impacts the road network efficiency. The problem consists on generate appropriate signal patterns by controlling the timing of the green/red light cycles at an intersection with the goal of optimally reduce congestion and the amount of time wasted stuck in traffic. This is a very complicated problem. This paper deals with the solution of the multi-traffic signal control problem for continuos-time Markov games under the expected average cost criterion. We consider three intersecting unidirectional roads that imply three possible traffic phases which alternate. This game describes a Poisson process where the cars leave the queue in the order they arrive. The optimization approach is applied to constant arrival flows λ and departure flows µ over short time periods (Δt) involved in our processes. A conflict appears when each signal controller tries to minimize its queue. The c??variable method is extended with a special restriction for continuous-time Markov chains to obtain the set of average optimal policies. The usefulness of the method is demonstrated empirically by an application example.
Non-recurrent traffic incidents (accidents, stalled vehicles and spilled loads) often bring about traffic congestion and even secondary accidents. Detecting and positioning them quickly and accurately has important significance for early warning, timely incident-disposal and speedy congestion-evacuation. This study proposes a video-based detecting and positioning method by analysing distribution characteristics of traffic states in a road segment. Each lane in the monitored segment is divided into a cluster of cells. Traffic parameters in each cell, including flow rate, average travel speed and average space occupancy, are obtained by detecting and tracking traffic objects (vehicles and spilled loads). On the basis of the parameters, traffic states in the cells are judged via a fuzzy-identification method. For each congested cell, a feature vector is constructed by taking its state together with states of its upstream and downstream neighbouring cells in the same lane. Then, a support vector machine classifier is trained to detect incident point. If a cell is judged to be corresponding to an incident point at least for two successive time periods, an incident is detected and its position is calculated based on the identity number of the cell. Experiments prove the efficiency and practicability of the proposed method.
The real-time traffic parameters are necessary to dynamic traffic light control at intersection due to the serious traffic congestion. In this paper, we describe an approach for the real-time vehicle queue length measurement in a video-based traffic monitoring system. It is built on the property of a modified local variance in video frames, which does not rely on any sort of motion detection. In addition, a shadow removal approach is also presented with a simplified LBP as well as this local variance. Experimental results show that the proposed approach can highly improve the performance of the queue length measurement in real time, and it can be efficiently applied in a variety of traffic scenes.
People across the world habitually turn to online social media to share their experiences, thoughts, ideas, and opinions as they go about their daily lives. These posts collectively contain a wealth of insights into how masses perceive their surroundings. Therefore, extracting people's perceptions from social media posts can provide valuable information about pertinent issues such as public transportation, emergency conditions, and even reactions to political actions or other activities. This paper proposes a novel approach to extract such perceptions from a corpus of social media posts originating from a given broad geographical region. The approach divides the broad region into a number of sub-regions, and trains language models over social media conversations within these sub-regions. Using Bayesian and geo-smoothing methods, the ensemble of language models can be queried with phrases embodying a perception. Discrete and continuous visualization methods represent the extent to which social media posts within the sub-regions express the query. The capabilities of the perception mining approach are illustrated using transportation-themed scenarios.
The traffic in urban areas is mainly regularized by traffic lights, which may contribute to the unnecessary long waiting times for vehicles if not efficiently configured. This inefficient configuration is unfortunately still the case in a lot of urban areas, where most of the traffic lights are based on a 'fixed cycle' protocol. To improve the traffic light configuration, this paper proposed monitoring system to be as an additional component (or additional subsystem) to the intelligent traffic light system, this component will be able to determine three street cases (empty street case, normal street case and crowded street case) by using small associative memory. The proposed monitoring system is working in two phases: training phase and recognition phase. The experiments presented promising results when the proposed approach was applied by using a program to monitor one intersection in Penang Island in Malaysia. The program could determine all street cases with different weather conditions depending on the stream of images, which are extracted from the streets video cameras. In addition, the observations which are pointed out to the proposed approach show a high flexibility to learn all the street cases using a few training images, thus the adaptation to any intersection can be done quickly.
A novel method for measurement of vehicle queue length is presented in this paper. Previous methods proposed by researchers for queue detection are based on video sequences. The method proposed here measures the queue length in signal-frame image acquired by a stationary camera, which avoids the effect of uncontrollable camera motions. Besides, considering the varying illumination, double-threshold method is used for image segmentation and flashlight is applied to the system for realizing the round-the-clock traffic status detect. The experiment results of queue length measure from images of real road scenes show the accuracy and reliability of the proposed method.
Based on video processing technology, this paper proposes a spatial-temporal Markov random field method. This method can track the position changes of vehicles queue tail and head at intersections in real time, which accurately describes the formation and dissipation of the queue. When the tracking algorithm is used, first we decided on an allocation of an initial label. Then through the spatial-temporal Markov random field, we refined and optimized the label's allocation. In the refining process, the algorithm should consider the piece's relation between time and space, and the Markov model can be used, and then a label can be assigned to a piece. Experimental results show the accuracy rate is more than 92% by real-time tracking of tail and head positions under varying weather conditions and levels of illumination.
In this paper, we present the design and analysis of an adaptive cost-effective discrete multitone transponder (DMT) using direct detection (DD) suitable for data center interconnections. Levin Campello margin adaptive (LC-MA) algorithm is applied to the transponder digital signal processing modules to enhance fiber chromatic dispersion (CD) resilience, while achieving high-data rate transmission. The bit error rate (BER) performance and the rate/distance adaptive capabilities of the proposed transponder have been numerically analyzed and compared to bandwidth variable uniform loading, taking into account the transmission impairments at the varying of the fiber length. Specifically, the performance of the designed transponder has been assessed from $20$ to $112$ Gb/s, extending the achievable reach at $50$ Gb/s beyond $80$ km of standard single mode fiber (SSMF). The numerical simulations have been compared with experimental results, evidencing good agreement in presence of transmission impairments.
There is a worldwide effort to apply 21st century intelligence to evolving our transportation networks. The goals of smart transportation networks are quite noble and manifold, including safety, efficiency, law enforcement, energy conservation, and emission reduction. Computer vision is playing a key role in this transportation
evolution. Video imaging scientists are providing intelligent sensing and processing technologies for a wide variety of applications and services. There are many interesting technical challenges including
imaging under a variety of environmental and illumination conditions, data overload, recognition and tracking of objects at high speed, distributed network sensing and processing, energy sources, as well
as legal concerns. This paper presents a survey of computer vision techniques related to three key problems in the transportation domain: safety, efficiency, and security and law enforcement. A broad review of the literature is complemented by detailed treatment of a few selected algorithms and systems that the authors believe represent the state-of-the-art.
This paper proposes a cellular automata-based solution of a two-dimensional binary classification problem. The proposed method is based on a two-dimensional, three-state cellular automaton (CA) with the von Neumann neighborhood. Since the number of possible CA rules (potential CA-based classifiers) is huge, searching efficient rules is conducted with use of a genetic algorithm (GA). Experiments show an excellent performance of discovered rules in solving the classification problem. The best found rules perform better than the heuristic CA rule designed by a human and also better than one of the most widely used statistical method: the k-nearest neighbors algorithm (k-NN). Experiments show that CAs rules can be successfully reused in the process of searching new rules.
This paper proposed a robust framework of detecting the real-time queuing and dissipation of a vehicle queue by two cameras, one fixed at the front of the stop line and the other somewhere behind the stop line, jointly monitoring the interested region with opposite and long-range views. Firstly, the position changes of the tail and head of a vehicle queue, which accurately describe the formation and dissipation of the queue, can be efficiently tracked in each camera at intersection during morning and evening rush hours, with a duplex flexible window fused with the Haar feature based AdaBoost cascade classifiers. Secondly, the data of these two cameras in this large-area outdoor traffic application are fused at decision level to improve the accuracy of the tracking, according to the tracking result in each camera. Then, the queue length and stop delay of vehicles can be calculated readily. Experiments show that the proposed method can detect the formation and dissipation of the queue under varying illumination in real time, and that the accuracy rate is about 90.24%. Therefore, this method can be further applied to traffic congestion monitoring and traffic signal controlling.
There is a world-wide effort to apply 21st century intelligence to
evolving our transportation networks. The goals of smart transportation
networks are quite noble and manifold, including safety, efficiency, law
enforcement, energy conservation, and emission reduction. Computer
vision is playing a key role in this transportation evolution. Video
imaging scientists are providing intelligent sensing and processing
technologies for a wide variety of applications and services. There are
many interesting technical challenges including imaging under a variety
of environmental and illumination conditions, data overload, recognition
and tracking of objects at high speed, distributed network sensing and
processing, energy sources, as well as legal concerns. This conference
presentation and publication is brief introduction to the field, and
will be followed by an in-depth journal paper that provides more details
on the imaging systems and algorithms.
Real-time qualitative road traffic data analysis is the cornerstone for any modern transport system. So far, most of the analysis is done manually and the use of image processing techniques for qualitative analysis is at its early stage. In this paper we describe novel image processing algorithms together with the results, which assign a qualitative description to a traffic scene. The qualitative description of a traffic scene can be used for controlling traffic lights and putting hazard signals on the road side, thereby warning drivers to slow down or direct them to alternative routes. We analyse a wider view of the path and evaluate the whole description of traffic status. To approach this, we considered two major parameters of traffic status: the percentage of road occupied by vehicles and the percentage of moving and stationary parts of this occupancy. The alogrithm developed for traffic analysis automatically divides the scene into a number of blocks, based on camera parameters and the number of lanes. This full frame image processing application requires a low-cost frame grabber and a Pentium-based computer system for on-line real-time operations.
In recent years, video-based Intelligent Transportation Systems (ITS) have been of major importance for enforcing traffic management policies. We propose a real-time and effective method for detecting vehicles from a sequence of traffic images taken by a single roadside mounted camera. The proposed algorithm includes three stages: first, extract moving object region from the current input image by background subtraction method, second, eliminate moving cast shadow which is often caused by moving vehicle and, at last, detect vehicle so that there can be a unique object associated with each vehicle. The proposed method has been tested on a number of monocular traffic-image sequences and the experimental results on the real-world videos show that the algorithm is effective and real-time. The correct rate of vehicle detection is higher than 90 percent, independent of environmental conditions.
By vehicle presence detection and movement analysis on the monitored region based on video processing technology, a real-time robust detection method of vehicle queue and dissipation is proposed in this paper. With a duplex flexible window fused with the Haar feature based AdaBoost cascade classifiers, the method can efficiently track the position changes of the tail and head of a vehicle queue at intersection during evening rush hour, which accurately describes the formation and dissipation of the queue. Then, the queue length and stop delay of vehicles can be calculated. Experiments showed that our method can real-time track the formation and dissipation of the queue during the evening rush hour when the illumination varies greatly from light to dark, and that the accuracy rate is about 91.54%. Therefore, this method can be further applied to traffic congestion monitoring and traffic signal controlling.
A novel vision-based method for real time extracting vehicle flow information on a road is proposed in this paper. Two new vision parameters, called as the contrast and luminance distortion ones respectively, are defined and employed to extract the vehicle flow information on the road. The analytical results show that the contrast distortion parameter of a video sequence can restrain the effects of the shadow interferences on the vehicle flow measurements, while the luminance distortion parameter is good for real-time background updating and supplement in order to enhance the accuracy of the measurements. In addition, the unsolved problem of the traditional vision-based methods, failing to distinguish two close consecutive vehicles, can be relaxed to a large extent by combining information from such two parameters. The experimental results from the different roads and vehicle flows under the different whether conditions show that the measurement accuracy of the proposed method is 97.27%.
This paper describes the development and use of a real-time machine vision system in the industrial production of mineral wool. Because of specific operating requirements and the very hostile environment the system employs unique hardware and software features.The main requirement is for real-time detection and control of the trajectory of molten rock as it falls onto a system of spinning wheels. The jet exhibits significant variation of its position with time, as a consequence of complex thermodynamic and hydrodynamic processes, resulting in a wide variation of quality and quantity of final product. Algorithms were developed to detect variations in the jet trajectory with time and this information is used to generate suitable control data.
Traffic engineers require various types of road traffic data to manage the traffic. The current sensors (inductive loops and axle sensors etc.) cannot collect all the data that are of interest to traffic engineers. Most important of all, if the data is to be collected at a different location, the installation of these equipments, which needs to be buried beneath the road, creates serious traffic disturbances. In order to overcome the above problems, many researchers have used vision-based system for collecting and analysing road traffic data. However, these techniques have not yielded good results due to various problems such as inefficiency of background updating, selection of a threshold value, change in ambient lighting etc. In this paper we describe a novel neural network and window-based image processing technique for road traffic applications. We use morphological edge detection techniques to detect vehicles. Once the vehicles have been detected, a back-propagation neural network is used for calculating various traffic parameters. This novel method has been implemented on a Pentium-based microcomputer system and the results are reported online in real-time. We also have compared our system with traditional image processing based systems and the results indicate that our proposed system provides better results than the traditional image processing based systems
Different types of road traffic parameters are required for modern road traffic control and management operations. Image processing based systems have been used to for collecting and analyzing road traffic data. However, these techniques have not yielded good results due to various problems such as inefficiency of background updating and selecting a threshold value, change in ambient lighting etc., which resulted in false detection of vehicles. In this paper we describe novel morphological edge detection and window based image processing technique for road traffic applications. This novel method has been implemented on a Pentium-based microcomputer system and the results are reported online in real-time. We also have compared our system with other traditional image processing based systems and the results indicate that our proposed system provides better results than the traditional image processing based systems
ResearchGate has not been able to resolve any references for this publication.