Vision-based real-time traffic accident detection

Abstract

Recently, traffic accident detection is becoming one of the interesting fields due to its tremendous application potential in Intelligent Transportation Systems. In this paper, we present a vision-based real time traffic accident detection method. We intend to extract foreground and background from video shots using the Gaussian Mixture Model (GMM) to detect vehicles; afterwards, the detected vehicles are tracked based on the mean shift algorithm. Then the three traffic accident parameters including the changes of the vehicles position, acceleration, and the direction of the moving vehicles are gathered to make the final accident decision. The experimental results on real video demonstrate the efficiency and the applicability of the proposed approach.

Realtime Accident Detection and Alarm Generation System Over IoT

“…It's very much challenging in devising a vision-based algorithm for the given task. If in practice it is concerned, then there are many challenging factors for the performance of traffic accident detection algorithms in computer vision [8][9][10]. Some of the challenging factors are such as different weather scenarios, traffic density, etc.…”

“…The framework should also send the position of the car to the car holder's phone number via Text messaging with use of a Gps module and a text messaging to the telematics provider console via GPRS. A further tracking approach has been introduced and then the accusers of traffic deaths are provided with faster medical assistance [10]. In order to improve the survival potential of passengers involved in car accidents, the project proposes also a prototype architecture [11].…”

“…A homogeneous approach is also suggested by [10] in which the scales are evaluated from the tracks of the in-motion vehicles acquired by base modeling with the use of Gaussian mixture model (GMM) and locating with the use of mean shift algorithm. For implementation and utility, this approach is simple and easy but it is not appropriate for deprived environments such as weather conditions or sudden change in traffic pattern because it displays only change in position and also speed parameters and it can lead to false alarm also if there is any unexpected movement of a vehicle.…”

Automatic Detection of Traffic Accidents from Video Using Deep Learning Techniques

“…The next group is based on methods related to machine learning and statistics [7,9,39,40]. Many of these have presented solutions using artificial neural networks [41][42][43], support vector machines [16,44], probabilistic neural networks [45], autoencoders [46], block clustering [47], Random Forest [15], pattern recognition [18], image processing techniques [48] and the Hidden Markov Model [49], among others.…”

“…This approach can present false alarms with the auto-occlusion of objects of interest in the video. Using the same type of video with first-person vision, [40] presents a method for the detection of traffic accidents consisting of three steps: vehicle detection, tracking, and attribute extraction. A mixed Gaussian model is used for the detection of the moving vehicle and, by means of the mean displacement algorithm, the localized vehicle is tracked.…”