Illegal U-turn Research Articles

Identifying Wrong-Way Driving Incidents from Regular Traffic Videos Using Unsupervised Trajectory-based Method

Currently, transportation agencies have implemented different wrong-way driving (WWD) detection systems based on loop detectors, radar detectors, or thermal cameras. Such systems are often deployed at fixed locations in urban areas or on toll roads. The majority of rural interchange terminals does not have real-time detection systems for WWD incidents. Portable traffic cameras are used to temporarily monitor WWD activities at rural interchange terminals. However, it has always been a time-consuming task to manually review those videos to identify WWD incidents. The objective of this study was to develop an unsupervised trajectory-based method to automatically detect WWD incidents from regular traffic videos (not limited by mounting height and angle). The principle of the method includes three primary steps: vehicle recognition and trajectory generation, trajectory clustering, and outlier detection. This study also developed a new subtrajectory-based metric that makes the algorithm more adaptable for vehicle trajectory classification in different road scenarios. Finally, the algorithm was tested by analyzing 357 h of traffic videos from 14 partial cloverleaf interchange terminals in seven U.S. states. The results suggested that the method could identify all the WWD incidents in the testing videos with an average precision of 80%. The method significantly reduced person-hours for reviewing the traffic videos. Furthermore, the new method could also be applied in detecting and extracting other kinds of abnormal traffic activities, such as illegal U-turns.

Self-Reported Habitual Practices and Perceptions of Young Drivers in Sri Lanka

Crashes on the road have become a significant socio-economic problem. Younger generations, who have lesser experience in driving, are at greater risks of facing road accidents. It is therefore important to identify driving practices and perceptions of young and inexperienced drivers at an early phase of exposure so that factors that improve safe driving can be identified.This study analysed 400 young and inexperienced drivers’ self-reported habitual practices and perceptions based on a questionnaire survey. The questionnaire focused on supervision of early driving, limitation made by parents, accident and traffic offenses by these youngsters as drivers as well as other habitual driving practices. Analyses were done using reliability statistics, inter-item correlation, likelihood ratio tests, and parameter estimates. The highest inter-item correlation value was 0.467 for the pair of overtaking vehicles in restricted areas and taking the chance to speed and run a yellow light when it is about to change to red. Driving after alcohol consumption, mobile phone usage, taking an illegal U-turn at restricted areas, non-use of signals when changing lanes and overtaking a slow driver from the left side were influenced non-use of seat-belts by young drivers. These habitual practices while driving highlighted the importance of early intervention to improve road safety.In conclusion, there is always one or more habitual driving practices that have affect the other driving habits of a young driver.

Adaptive density based data mining technique for detection of abnormalities in traffic video surveillance

Detection of abnormal events in a traffic scene is a highly challenging task due to vast field of view, continuous stream of video data, various object interactions and complex events in Video Surveillance. Hence, this research proposes novel schemes using machine learning approach to detect abnormal events such as illegal U-turn, presence of pedestrian in driving region, wrong side driving and frequent lane change. Recently, Density Based Spatial Clustering of Applications with Noise (DBSCAN) is a popular method that has been used for clustering the trajectory datasets. The existing Density Based Clustering approach used for Abnormal detection in traffic scene uses random selection of cluster radius (Eps) and minimum points (minpts) needed to form a cluster. This random selection is time consuming and inefficient clustering results in accuracy reduction in abnormal detection. So, Adaptive Density based Spatial Clustering of Applications with Noise (ADBSCAN) is proposed for the detection of abnormal events based on spatial temporal information relating to individual objects which determines the optimal values for the cluster radius (Eps) using the slope calculation of the K-d plot. Gaussion Mixture Model (GMM) is used for obtaining the moving foreground regions and region-based tracking is used for the identification of the objects in successive frames. The centroid of the region is calculated using image moments. If there is an occlusion between the vehicles then vehicle identification number (Id no) is used to differentiate them. The main advantage in this technique is clustering/labelling the normal pattern without the help of manual intervention. The effectiveness of ADBSCAN is experimentally evaluated using a real time benchmark video traffic dataset and it found that it gives better accuracy in detecting anomalies than the state-of-the-art techniques.

Red light running motorcyclists at signalized intersection in Malaysia: An empirical study

This study investigates factors associated with red light running motorcycle (RLR-MC) behavior during their approach and crossing of signalized 3-leg and 4-leg intersections along majors in Malaysia. Twenty-seven (27) intersections with Pre-timed traffic light (PTTL) and Actuated traffic light (ATL) were selected and observed during peak and off peak hour period. In general, the average rate of RLR-MC was 3.61%, with the highest rate of RLR-MC recorded was 22%, while the lowest rate was 0.6%. Observations have shown three-ways of RLR-MC approached the signalized intersection, i.e. (1) approaching the SI with Weaving/Lane Splitting, (2) approaching the SI from center of the lane, and (3) approaching the SI from the left side/the shoulder. There are three ways RLR-MC crossing the signalized intersection, i.e. (1) by Illegal maneuver (i.e. Illegal U-turn, contra-flow, prohibited left-turn), (2) by stopping first at or before the stop line, and (3) without stopping before the stop line. ‘Crossing the SI without stopping at the stop line’ is the majority with 51.15% from total observation. It is recommended that SI with high RLR-MC be equipped with adequate traffic island, replace SI signal type to actuated traffic signal and installing traffic light pole instead of gantry.

Trajectory based abnormal event detection in video traffic surveillance using general potential data field with spectral clustering

Detection of abnormal trajectories in a traffic scene is an important problem in Video Traffic Surveillance (VTS). Recently, General Potential Data Field (GPDf)-based trajectory clustering scheme has been adopted for detecting abnormal events such as illegal U-turn, wrong side and unusual driving behaviors and it uses spatial and temporal attributes explicitly. The concept of data field is used to discover the relation between the spatial points in data-space and grouping them into clusters based on their mutual interaction. Existing methodologies related to potential data field-based clustering have certain limitations such as pre-defined cluster size, non-effective cluster center identification, and limitation in range estimation using isotropic impact factor (h) which leads to inaccurate results. In order to address the above-mentioned issues, this paper proposes an efficient anomaly detection scheme based on General Potential Data field with Spectral Clustering (GPDfSC). The proposed GPDfSC scheme utilizes potential data field technique along with spectral clustering for effective identification of abnormalities. The Limitation in impact factor(h) is overcome by using anisotropic impact parameter Bmat. Further, Bayesian Decision theory is used to classify the events as normal or abnormal. The proposed scheme is implemented in real time using GPU and from the results it is found that it gives 12% better accuracy in detecting abnormalities than the state of art technique.

Investigating and Modeling the Illegal U-Turn Violations at Medians of Limited Access Facilities

Illegal U-turns on freeways and toll roads are risky maneuvers that sometimes result in the turning vehicles causing various types of collisions or disturbances to approaching traffic. These illegal U-turn maneuvers can occur at traversable grass medians and emergency crossovers. Limited literature was found regarding the impact of illegal U-turns on these facilities. Therefore, to understand the roadway and median characteristics that could influence drivers’ propensity to commit illegal U-turns, a sequential modeling methodology was adopted. This methodology combined a Poisson regression model with a Least Absolute Shrinkage and Selection Operator (LASSO) regression procedure to predict the cited violations at traversable median segments. Additionally, a logistic regression model was developed to predict the probability of a cited violation at official use only emergency crossovers. These models included illegal U-turn citations and crashes for the Orlando and Miami metropolitan areas in Florida from 2011 to 2016. The findings indicated that the average distance between access points, median width, speed limit, segment length, and distance to nearest segment were significant in predicting cited violations at traversable medians. Furthermore, the distance to the nearest interchange, distance to the nearest adjacent crossover, and median width were significant in predicting the probability of a cited violation occurring at an emergency crossover. This study helps agencies to predict the locations of illegal U-turn violations and to prioritize roadways for possible treatment to minimize the potential risk of head-on or other collisions due to illegal U-turn events.

Risky behaviors of taxi drivers in Bandar Abbas, Iran.

Background and aimIt has been revealed that taxi drivers break more traffic rules than ordinary drivers. Such risky behaviors include stopping at prohibited areas and sudden change of direction. The present study aimed to determine the Risky Behaviors of Taxi Drivers in Bandar Abbas, IranMethodsIn this cross-sectional study, 184 taxi drivers were randomly selected from eight taxi stations located at different parts of Bandar Abbas city in 2016. Taxi drivers’ risky behaviors were evaluated via a 20-item questionnaire. Data were analyzed by SPSS version 19, using descriptive statistics and independent-samples t-test. The p-values less than 0.05 were considered as statistically significant.ResultsThe mean age of the drivers was 45.1 (±11.1) years. The mean of their occupational experience was 18.7 (±10.8) years. The risky behaviors which showed the highest frequency were respectively, failure to use signal-lights, driving too close to the cars in front, refusing to drive within the lanes and erratic lane changing.The lowest frequency belonged to running a red light, ignoring ‘no entry’ signs and taking illegal U-turns. Risky driving behaviors were shown to be significantly more prevalent among drivers with previous experience of crashes or tickets than drivers with no such experiences (p<0.01).ConclusionIncreasing the role of police supervision for the strict implementation of driving laws, and modification of the drivers’ behavior and implementation of periodic training programs on drivers’ safety issues can be considered for reducing taxi drivers’ unsafe behaviors.

Real-time video processing for traffic control in smart city using Hadoop ecosystem with GPUs

Advancement in technology causes the rise in smart systems. City authorities want to make their cities smarter by making an intelligent decision at real time without the involvement of humans. Monitoring and controlling city traffic is one of the major challenges faced by the authorities. These days, city traffic is monitored by static network cameras deployed on few places of highways. Most of the vehicles are also equipped with cameras to store the videos as a black box. However, monitoring and controlling city traffic by using these thousands of cameras produce an overwhelming amount of high-speed videos, which is challenging to process at real time. Therefore, in this paper, we proposed a system to control city traffic by identifying illegal traffic behaviour, such as illegal U-turn, through continuous monitoring of city traffic. The continuous city traffic is monitored by the network static cameras placed on the road as well as by all the vehicles’ cameras. An architecture is proposed to handle high-speed vast volume of real-time videos efficiently. For that, the two-level parallelism is achieved with the combination of Hadoop and graphics processing unit (GPU) while processing each frame using parallel environment of Hadoop and each block of a frame using GPU. MapReduce Hadoop programming paradigm is not suitable for real-time processing. Thus, we proposed a parameter calculation algorithm that is equivalent to MapReduce mechanism for image processing while dividing the images/frames into fixed-size blocks. We analyzed the city road traffic, which is collected by static cameras placed on various roads and also by vehicles’ cameras while running on the road. Later, the illegal traffic behaviour are recognized, e.g. illegal U-turn, drunken drive, zig-zag drive, over-speed, etc. Finally, the efficiency of the designed system and algorithms are tested by considering the overall running time and system’s throughput with respect to video duration as well as the number of frames. The findings indicate that the proposed architecture with GPU-based algorithm over the Hadoop system perform extraordinary.

Dangerous driving behavior detection using video-extracted vehicle trajectory histograms

ABSTRACTDangerous driving behavior detection can be used in video surveillance systems to identify dangerous patterns, such as Abrupt Double Lane Change (ALC), Retrograde Driving (RD), and Illegal U-Turn (IT), for traffic design, traffic management, and law enforcement. The purpose of this study is to develop a detection method of dangerous driving behavior based on video surveillance. First, a novel method named trajectory histogram is proposed. A set of trajectory histograms (e.g., control points histogram and velocity histogram) is constructed to represent vehicle motion. Then, a frequently used feature selection method named Minimum Redundancy and Maximum Relevance (mRMR) is introduced to evaluate the most representative trajectory histograms for dangerous driving behavior detection. In addition, a hybrid algorithm, Particle Swarm Optimization-Support Vector Machine (PSO_SVM), is then developed to identify dangerous driving behavior. To validate the performance and effectiveness of the proposed method, several experiments are conducted. The results show that mRMR is better than other representative methods, namely Conditional Mutual Information Maximization (CMIM), Mutual Information Maximization (MIM), and ReliefF, for evaluating the most representative trajectory histograms. Based on the most representative trajectory histograms, the detection accuracy rate of dangerous driving behavior using PSO_SVM is superior to those of the most frequently used classifiers—Naïve Bayesian Classifier (NBC), k-Nearest Neighbor (kNN), and C4.5 decision tree. In addition, we find that the proposed method outperforms the two common approaches for dangerous driving behavior detection in video surveillance systems. Therefore, the proposed method can be widely applied to detect dangerous driving behavior in video surveillance systems.

고위험군 운전자의 운행행태 판단기술 개발

교통사고의 발생의 주요 원인의 대부분은 운전자의 법규 위반 행위이며, 특히 과속 및 불법유턴은 대형 사고로 이어질 잠재적 위험성을 내포한다. 사업용 차량의 경우 불법유턴 위험성에 대한 지각이 부족하고, 단속장비 및 감시 인력 또한 부족한 실정이다. 이에 본 논문에서는 사업용 운전자의 불법유턴 위반행위를 예방하고, 안전운행 습관을 유도하고자, 맵 연동기반 운행행태 판단 기술을 개발에 대해 소개한다. 이를 위해 전국 총 26만개 링크자료 중 23,782 건의 유턴 허용 정보와 14만 6천 건의 제한속도 정보를 수집하였다. 또한 운행행태 분석을 위해 판단알고리즘을 구축하였으며, 최종적으로 안드로이드 기반 애플리케이션을 개발하였다. 개발된 애플리케이션의 맵 매칭, 운행행태 판단 및 기타 데이터 산출의 정확도를 평가하기 위해 테스트를 수행하였으며, 분석결과 맵 매칭은 86%, 운행행태 판단 정확도는 83%로 평가되었다. 또한 정보출력 표출에 대한 정확도는 모두 100%로 평가되었다. 향후 과제로서는 기술의 정확도 향상을 위해 모니터링 시스템 설계가 필요할 것으로 판단되며, 개발된 기술을 활용할 수 있는 서비스 시나리오의 검토가 필요하다고 판단된다. Driving behaviors such as speeding and illegal u-turn which violate traffic rules are main causes of car accidents, and they can lead to serious accidents. Bus drivers are less aware of dangers of illegal u-turn, and infrastructures such as traffic enforcement equipment and watchmen are deficient. This research aims to develop technology for estimating driving behaviors based on map-matching in order to prevent illegal u-turns. For this research, 23,782 of u-turn permit data and 146,000 of speed limit data are collected nationwide, and an estimation algorithm is built with these data. Then, an application based on android is developed, and finally, tests are conducted to assess the accuracy in data computations and GPS data map-matching, and to extrapolate driving behavior. As a result of the tests, the accuracy results in the map-matching is 86% and the assessment of driving behavior is 83%, while the display of the data output yielded 100% accuracy. Additional research should focus on improvement in accuracy through the development of a robust monitoring system, and study of service scenarios for technology application.

옵티컬 플로우 분석을 통한 불법 유턴 차량 검지

Today, Intelligent Vehicle Detection System seeks to reduce the negative factors, such as accidents over to get the traffic information of existing system. This paper proposes detection algorithm for the illegal U-turn vehicles which can cause critical accident among violations of road traffic laws. We predicted that if calculated optical flow vectors were shown on the illegal U-turn path, they would be cause of the illegal U-turn vehicles. To reduce the high computational complexity, we use the algorithm of pyramid Lucas-Kanade. This algorithm only track the key-points likely corners. Because of the high computational complexity, we detect center lane first through the color information and progressive probabilistic hough transform and apply to the around of center lane. And then we select vectors on illegal U-turn path and calculate reliability to check whether vectors is cause of the illegal U-turn vehicles or not. Finally, In order to evaluate the algorithm, we calculate process time of the type of algorithm and prove that proposed algorithm is efficiently.

Evaluation of Innovative Bicycle Facilities in Washington, D.C.

Two innovative bicycle facilities were installed in Washington, D.C., during 2010 by the District Department of Transportation: buffered center median bicycle lanes on Pennsylvania Avenue, NW, and a two-way cycle track on 15th Street, NW. Both facilities included dedicated road space with buffers between bicyclists and motor vehicles, signal control, and signs and pavement markings. The facilities were designed to increase safety, comfort, and convenience for cyclists. This paper presents the results of a comprehensive study that evaluated the facilities to understand how well they worked for cyclists, motorists, and pedestrians in terms of safety and level of service (LOS) and how they affected behavior and attitudes. The study found that bicycle LOS improved and that bicycle volume on those corridors nearly quadrupled, well above the rate of citywide bicycle use. Motor vehicle LOS was largely unaffected. Signal progression was mixed for bicyclists: contraflow travel on the one-way portion of 15th Street and travel against the dominant direction on Pennsylvania Avenue showed the worst travel time performances. Although bicycle crashes increased on both facilities, the crash rate remained similar on 15th Street to what it had been previously. Pennsylvania Avenue saw a higher crash rate, mainly as a result of illegal U-turns across the bike lanes. In surveys of all travelers on the corridor (i.e., bicyclists, motorists, pedestrians) and nearby residents, the perception of the lanes in general was positive for all users, and the lanes were seen as a positive addition to the community.

Video Anomaly Identification

This article describes a family of unsupervised approaches to video anomaly detection based on statistical activity analysis. Approaches based on activity analysis provide intriguing possibilities for region-of-interest (ROI) processing since relevant activities and their locations are detected prior to higher-level processing such as object tracking, tagging, and classification. This strategy is essential for scalability of video analysis to cluttered environments with a multitude of objects and activities. Activity analysis approaches typically do not involve object tracking, and yet they inherently account for spatiotemporal dependencies. They are robust to clutter arising from multiple activities and contamination arising from poor background subtraction or occlusions. They can sometimes also be employed for fusing activities from multiple cameras. We illustrate successful application of activity analysis to anomaly detection in various scenarios, including the detection of abandoned objects, crowds of people, and illegal U-turns.

On Chemo Brain, Citizens' Arrests, and the Development of Kittens

My brain has a mind of its own. Most of the time, give it commands in much the same way Captain Kirk gave commands to the Enterprise computer. Computer, plot a course for Alpha Centauri. Without hesitation, the computer replies, Course plotted. In my case, the commands have always been far more mundane, Give me five more ways can teach John about regrouping. Or Give me three other uses for egg cartons. So, on the whole, I've always felt pretty confident that was in control of my brain. Until recently. For several months, paid people to regularly inject large amounts of poison into my veins, poison designed to kill as many newly forming cancer cells as possible--without killing me in the process. Chemo's side effects are numerous and varied. So no matter how felt, how my body responded, or what strange occurrences reported to my oncologist--and there were strange things to report--my oncologist simply responded, Yes, that happens with chemo. Over time, became pretty sure that if reported I'd grown a second head, his reply would have been the same. As my body mounted its forces to fight the poison, got tired--very tired--and spent a lot of time lying on the couch drifting in and out of sleep as the TV droned on in the background. That's when my brain began to assert its control. The doctor explained that my body's efforts to expel the poison meant less oxygen flowed to my brain. As a result, my brain, usually busy with my many commands, was suddenly unoccupied. And, I've learned, brains left with too much time on their hands can get into trouble. It happened like this. One afternoon drifted off to sleep while watching the old Andy Griffith Show. In this episode, overzealous deputy Barney Fife gets into a row with n'er-do-well service station operator Gomer Pyle. When Barney tickets Gomer for a traffic violation, Gomer protests. Later, when Gomer sees Barney make an illegal U-turn, he rushes into the street yelling, Citizen's arrest! Gomer's voice has remained etched in my memory since first heard him utter those words years ago. Babies and Reading Drifting back to consciousness some time later, realized Mayberry had been replaced by an infomercial touting a new reading program for babies. This program promised to revolutionize how reading is taught. Using an animated illustration and the word clap, the program's developer pointed out that Baby would hear the word, see the word, and do the word. When the baby in the illustration claps, a light blinks in Baby's head, scientifically proving his argument's validity. The plan would be great, mused, if reading involved just calling out the names of concrete items and action words. But how would this program teach a baby to read the word the, wondered. I've seen such programs before, and suspect I'll see them again. They're created, think, by well-intentioned folk but marketed by charlatans who play on the emotions of concerned parents. Even as thought this, heard the words, Citizen's arrest! coming from my mouth. But had not commanded my brain to say those words. My brain was acting on its own, and it--my brain--seemed seriously out of control. Google 'police surplus stun guns', my brain demanded. We'll need some means of enforcement even for a citizen's arrest. was horrified! You can't arrest the guy for trying to sell parents a miracle teaching program, no matter how much of a charlatan he is, explained to my vigilante brain. I mean, it's just advertising. It's the same as coal companies touting 'clean coal' or claims that a Taco Bell drive-through diet is a good weight-loss plan. It's a buyer-beware world. Besides, these things come and go. …

Traffic monitoring and accident detection at intersections

We have developed an algorithm, referred to as spatio-temporal Markov random field, for traffic images at intersections. This algorithm models a tracking problem by determining the state of each pixel in an image and its transit, and how such states transit along both the x-y image axes as well as the time axes. Our algorithm is sufficiently robust to segment and track occluded vehicles at a high success rate of 93%-96%. This success has led to the development of an extendable robust event recognition system based on the hidden Markov model (HMM). The system learns various event behavior patterns of each vehicle in the HMM chains and then, using the output from the tracking system, identifies current event chains. The current system can recognize bumping, passing, and jamming. However, by including other event patterns in the training set, the system can be extended to recognize those other events, e.g., illegal U-turns or reckless driving. We have implemented this system, evaluated it using the tracking results, and demonstrated its effectiveness.