Wrongway Drivers Detection Based on Optical Flow

Monteiro, Gonçalo; Ribeiro, M.; Marcos, João; Batista, Jorge

doi:10.1109/icip.2007.4379785

Cited by 28 publications

(20 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, extensions related to detecting anomalous traffic behaviour, such as detecting the presence of vehicles moving in the wrong direction (cf. [31]), is another interesting direction for future research. Third, attention in the current system has been limited to vehicle-based scene elements.…”

Section: Discussion and Summarymentioning

confidence: 98%

“…For example, statistics of optical flow taken over the roadway have been used to characterize traffic flow [41,27,32,31,26]. A drawback of these approaches is that extracting reliable measurements of flow is difficult in traffic scenarios due to environmental conditions and are subject to noise in cases where there are many vehicles in the scene.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Classification of traffic video based on a spatiotemporal orientation analysis

Derpanis

Wildes

2011

2011 IEEE Workshop on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

This paper describes a system for classifying traffic congestion videos based on their observed visual dynamics. Central to the proposed system is treating traffic flow identification as an instance of dynamic texture classification. More specifically, a recent discriminative model of dynamic textures is adapted for the special case of traffic flows. This approach avoids the need for segmentation, tracking and motion estimation that typify extant approaches. Classification is based on matching distributions (or histograms) of spacetime orientation structure. Empirical evaluation on a publicly available data set shows high classification performance and robustness to typical environmental conditions (e.g., variable lighting).

show abstract

Section: Discussion and Summarymentioning

confidence: 98%

Section: Related Workmentioning

confidence: 99%

Classification of traffic video based on a spatiotemporal orientation analysis

Derpanis

Wildes

2011

2011 IEEE Workshop on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

show abstract

“…Since our task in focus requires a minimal error rate, usage of optical flow as a detector could be unacceptable due to the changing lighting and weather conditions. Monteiro et al [8] described optical flow-based wrong-way driver detection. Their proposed system was separated into three main stages: learning, detection, and validation.…”

Section: Optical Flowmentioning

confidence: 99%

Real-Time, Deep Learning Based Wrong Direction Detection

2020

View full text Add to dashboard Cite

In this paper, we develop a real-time intelligent transportation system (ITS) to detect vehicles traveling the wrong way on the road. The concept of this wrong-way system is to detect such vehicles as soon as they enter an area covered by a single closed-circuit television (CCTV) camera. After detection, the program alerts the monitoring center and triggers a warning signal to the drivers. The developed system is based on video imaging and covers three aspects: detection, tracking, and validation. To locate a car in a video frame, we use a deep learning method known as you only look once version 3 (YOLOv3). Therefore, we use a custom dataset for training to create a deep learning model. After estimating a car’s position, we implement linear quadratic estimation (also known as Kalman filtering) to track the detected vehicle during a certain period. Lastly, we apply an “entry-exit” algorithm to identify the car’s trajectory, achieving 91.98% accuracy in wrong-way driver detection.

show abstract

“…Another approach for incident detection is to use GMM for background updates before implementing object detection (Monteiro et al 2007). While traditional detection methods analyze the data from loop detection systems at two locations, or generate traffic information based on the neural network, the GMM approach mostly adapts well to outdoor environments while updating the background, and is now used more often than the Kalman Filter method.…”

Section: Literature Reviewmentioning

confidence: 99%

Gaussian background mixture model based automatic incident detection system for real-time tracking

OhJutaek¹,

MinJoonyoung²

2011

Can. J. Civ. Eng.

View full text Add to dashboard Cite

The image-based incident detection system is capable of not only replacing the loop detector, which has limited management and operation functions, but can also record the sequential conditions before and after a traffic accident. Thus, it is possible to analyze its mechanisms objectively using this data. In this study, the researchers developed a reliable video image based accident detection system with a high detection rate and low error rates. The proposed accident detection algorithm in this study provides the preliminary judgment of potential accident by detecting stopped objects using the Gaussian Mixture Model. Afterwards, it measures the traces of vehicles, the speed variance, and the occupancy per detecting area to propose an algorithm that makes the final accident decision. The proposed algorithm performs accident detection by extracting the stopped objects based on the video image, the trajectory movement of the trailing vehicle, and the variance of speed and traffic volume in the detection area. Thus, it can minimize false detections and maximize the detection rate, making it possible to accurately interpret an accident site and the circumstances surrounding it. Moreover, it is advantageous that the detection rate does not decline under bad weather conditions such as cloudy, rainy, foggy, or snowy.Résumé : Un système de détection d'accidents basé sur des images peut non seulement remplacer le détecteur à boucle d'induction, lequel a des fonctions de gestion et d'opération limitées, mais peut également enregistrer les conditions séquen-tielles avant et après un accident de la circulation. Il est donc possible d'analyser ses mécanismes de manière objective en utilisant ces données. Les chercheurs de la présente étude ont développé un système de détection des accidents fiable, basé sur les images vidéo, et présentant un taux élevé de détection et de faibles taux d'erreur. L'algorithme de détection des accidents proposé dans cette étude fourni une évaluation préliminaire des accidents potentiels en détectant les objets arrêtés par l'utilisation du modèle de mélange gaussien. Puis, il mesure les traces des véhicules, la variation de la vitesse et l'occupation de la zone de détection afin de proposer un algorithme qui rend la décision finale d'accident. L'algorithme proposé détecte les accidents en extrayant les objets arrêtés selon l'image vidéo, la trajectoire du véhicule arrière, ainsi que la variation de la vitesse et le volume de circulation dans la zone de détection. Il peut donc minimiser les fausses détections et maximiser le taux de détection, rendant possible l'interprétation précise d'un site d'accident et les circonstances associées. De plus, il est avantageux que le taux de détection ne s'amenuise pas sous des conditions climatiques défavorables, telles que nuages, pluie, brouillard ou neige.

show abstract

Wrongway Drivers Detection Based on Optical Flow

Cited by 28 publications

References 4 publications

Classification of traffic video based on a spatiotemporal orientation analysis

Classification of traffic video based on a spatiotemporal orientation analysis

Real-Time, Deep Learning Based Wrong Direction Detection

Gaussian background mixture model based automatic incident detection system for real-time tracking

Contact Info

Product

Resources

About