Vehicle Crash Prediction using Vision

Haris, Muhammad; Moin, Minhaj Ahmed; Rehman, Farooq Abdul; Farhan, Muhammad

doi:10.1109/ciss50987.2021.9400257

Cited by 8 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Vision-based detection for accident or special traffic event detection has been a major research focus in the field, with most studies relying on surveillance cameras [2][3][4] as an infrastructure to generate vehicle trajectory data and detect [3] accidents or predict [4] any possible accident moments. Arceda et al [6] proposed a framework for the real-time detection of vehicle-to-vehicle accidents in the surveillance camera system.…”

Section: System-wise Methodologiesmentioning

confidence: 99%

“…Consequently, structured roads such as highways are expected to evolve to align with this new paradigm in transportation and changes in government policy, through the integration of real-time, automated traffic event diagnosis systems. For this infrastructure system to be realized, most automobile accident detection methods have been developed for third-party perspectives captured by static cameras [1][2][3][4][5][6], such as surveillance cameras. These systems benefit from their birds-eye view and ability to easily track vehicle trajectories, allowing for a comprehensive overview of traffic flow.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Special Traffic Event Detection: Framework, Dataset Generation, and Deep Neural Network Perspectives

Lee,

Noh

et al. 2023

Sensors

View full text Add to dashboard Cite

Identifying early special traffic events is crucial for efficient traffic control management. If there are a sufficient number of vehicles equipped with automatic event detection and report gadgets, this enables a more rapid response to special events, including road debris, unexpected pedestrians, accidents, and malfunctioning vehicles. To address the needs of such a system and service, we propose a framework for an in-vehicle module-based special traffic event and emergency detection and safe driving monitoring service, which utilizes the modified ResNet classification algorithm to improve the efficiency of traffic management on highways. Due to the fact that this type of classification problem has scarcely been proposed, we have adapted various classification algorithms and corresponding datasets specifically designed for detecting special traffic events. By utilizing datasets containing data on road debris and malfunctioning or crashed vehicles obtained from Korean highways, we demonstrate the feasibility of our algorithms. Our main contributions encompass a thorough adaptation of various deep-learning algorithms and class definitions aimed at detecting actual emergencies on highways. We have also developed a dataset and detection algorithm specifically tailored for this task. Furthermore, our final end-to-end algorithm showcases a notable 9.2% improvement in performance compared to the object accident detection-based algorithm.

show abstract

Section: System-wise Methodologiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Special Traffic Event Detection: Framework, Dataset Generation, and Deep Neural Network Perspectives

Lee,

Noh

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Early research pipelines in Vision-TAA with object location association borrow the trajectory prediction models and follow the pipeline of detection-tracking-trajectory predictionaccident determination [95]. In the meantime, early accident prediction research focuses on the surveillance fields, and the works follow the direct physical distance computation for accident prediction on the predicted trajectories.…”

Section: Vision-taa With Object Trajectory Predictionmentioning

confidence: 99%

Traffic Accident Detection via Self-Supervised Consistency Learning in Driving Scenarios

Fang

Qiao²,

Bai³

et al. 2022

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

Traffic accident detection and anticipation is an obstinate road safety problem and painstaking efforts have been devoted. With the rapid growth of video data, Vision-based Traffic Accident Detection and Anticipation (named Vision-TAD and Vision-TAA) become the last one-mile problem for safe driving and surveillance safety. However, the long-tailed, unbalanced, highly dynamic, complex, and uncertain properties of traffic accidents form the Out-of-Distribution (OOD) feature for Vision-TAD and Vision-TAA. Current AI development may focus on these OOD but important problems. What has been done for Vision-TAD and Vision-TAA? What direction we should focus on in the future for this problem? A comprehensive survey is important. We present the first survey on Vision-TAD in the deep learning era and the first-ever survey for Vision-TAA. The pros and cons of each research prototype are discussed in detail during the investigation. In addition, we also provide a critical review of 31 publicly available benchmarks and related evaluation metrics. Through this survey, we want to spawn new insights and open possible trends for Vision-TAD and Vision-TAA tasks.

show abstract

“…Therefore, artificial intelligence (AI) algorithms should be used by the machine to perform a problemsolving function. In transport, intelligent systems based on machine vision are used to solve such problems as: road safety analysis [19], intelligent traffic control at intersections [20], objects tracking [21], obstacle detection and navigation [22], crash prediction [23] etc.…”

Section: Introductionmentioning

confidence: 99%

Car Safety Support System on the Base of Data Mining Algorithm

Żabiński¹,

Shubenkova²,

Makarova³

et al. 2021

Proceedings of the Second Conference on Sustainable Development: Industrial Future of Territories (IFT 2021)

View full text Add to dashboard Cite

According to the target of the 11 th Sustainable Development Goal, to reach sustainability in cities, we have to ensure the transport systems' safety. Since safety driver assistance systems play a critical role both in averting crashes and reducing the likelihood of serious injury, we suggest a Safety Support System based on machine vision and a data mining algorithm that could be used for application to decision module of fully-or semi-autonomous vehicles. The created system was tested on a simple collection of photos from a Polish two-lane road. The novelty of this system is concluded in our EAV (Entity-Attribute-Value) model based algorithm that, while providing the same level of accuracy, works faster by reducing the number of analyzed attributes.

show abstract

Vehicle Crash Prediction using Vision

Cited by 8 publications

References 12 publications

Special Traffic Event Detection: Framework, Dataset Generation, and Deep Neural Network Perspectives

Special Traffic Event Detection: Framework, Dataset Generation, and Deep Neural Network Perspectives

Traffic Accident Detection via Self-Supervised Consistency Learning in Driving Scenarios

Car Safety Support System on the Base of Data Mining Algorithm

Contact Info

Product

Resources

About