Innovative Approaches to Traffic Anomaly Detection and Classification Using AI
Video anomaly detection plays a crucial role in intelligent transportation systems by enhancing urban mobility and safety. This review provides a comprehensive analysis of recent advancements in artificial intelligence methods applied to traffic anomaly detection, including convolutional and recurre...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/10/5520 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849327444781170688 |
|---|---|
| author | Borja Pérez Mario Resino Teresa Seco Fernando García Abdulla Al-Kaff |
| author_facet | Borja Pérez Mario Resino Teresa Seco Fernando García Abdulla Al-Kaff |
| author_sort | Borja Pérez |
| collection | DOAJ |
| description | Video anomaly detection plays a crucial role in intelligent transportation systems by enhancing urban mobility and safety. This review provides a comprehensive analysis of recent advancements in artificial intelligence methods applied to traffic anomaly detection, including convolutional and recurrent neural networks (CNNs and RNNs), autoencoders, Transformers, generative adversarial networks (GANs), and multimodal large language models (MLLMs). We compare their performance across real-world applications, highlighting patterns such as the superiority of Transformer-based models in temporal context understanding and the growing use of multimodal inputs for robust detection. Key challenges identified include dependence on large labeled datasets, high computational costs, and limited model interpretability. The review outlines how recent research is addressing these issues through semi-supervised learning, model compression techniques, and explainable AI. We conclude with future directions focusing on scalable, real-time, and interpretable solutions for practical deployment. |
| format | Article |
| id | doaj-art-3572adc2e09541e4a8f1ed00bea069e2 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-3572adc2e09541e4a8f1ed00bea069e22025-08-20T03:47:53ZengMDPI AGApplied Sciences2076-34172025-05-011510552010.3390/app15105520Innovative Approaches to Traffic Anomaly Detection and Classification Using AIBorja Pérez0Mario Resino1Teresa Seco2Fernando García3Abdulla Al-Kaff4Autonomous Mobility and Perception Lab (AMPL), Departamento de Ingeniería de Sistemas y Automática, Universidad Carlos III de Madrid, Av. de la Universidad, 30, 28911 Madrid, SpainAutonomous Mobility and Perception Lab (AMPL), Departamento de Ingeniería de Sistemas y Automática, Universidad Carlos III de Madrid, Av. de la Universidad, 30, 28911 Madrid, SpainTechnological Institute of Aragón, Calle Maria de Luna, 7-8, 50018 Zaragoza, SpainAutonomous Mobility and Perception Lab (AMPL), Departamento de Ingeniería de Sistemas y Automática, Universidad Carlos III de Madrid, Av. de la Universidad, 30, 28911 Madrid, SpainAutonomous Mobility and Perception Lab (AMPL), Departamento de Ingeniería de Sistemas y Automática, Universidad Carlos III de Madrid, Av. de la Universidad, 30, 28911 Madrid, SpainVideo anomaly detection plays a crucial role in intelligent transportation systems by enhancing urban mobility and safety. This review provides a comprehensive analysis of recent advancements in artificial intelligence methods applied to traffic anomaly detection, including convolutional and recurrent neural networks (CNNs and RNNs), autoencoders, Transformers, generative adversarial networks (GANs), and multimodal large language models (MLLMs). We compare their performance across real-world applications, highlighting patterns such as the superiority of Transformer-based models in temporal context understanding and the growing use of multimodal inputs for robust detection. Key challenges identified include dependence on large labeled datasets, high computational costs, and limited model interpretability. The review outlines how recent research is addressing these issues through semi-supervised learning, model compression techniques, and explainable AI. We conclude with future directions focusing on scalable, real-time, and interpretable solutions for practical deployment.https://www.mdpi.com/2076-3417/15/10/5520anomaly detectionmachine learningTransformersartificial intelligencecomputer visionroad safety |
| spellingShingle | Borja Pérez Mario Resino Teresa Seco Fernando García Abdulla Al-Kaff Innovative Approaches to Traffic Anomaly Detection and Classification Using AI Applied Sciences anomaly detection machine learning Transformers artificial intelligence computer vision road safety |
| title | Innovative Approaches to Traffic Anomaly Detection and Classification Using AI |
| title_full | Innovative Approaches to Traffic Anomaly Detection and Classification Using AI |
| title_fullStr | Innovative Approaches to Traffic Anomaly Detection and Classification Using AI |
| title_full_unstemmed | Innovative Approaches to Traffic Anomaly Detection and Classification Using AI |
| title_short | Innovative Approaches to Traffic Anomaly Detection and Classification Using AI |
| title_sort | innovative approaches to traffic anomaly detection and classification using ai |
| topic | anomaly detection machine learning Transformers artificial intelligence computer vision road safety |
| url | https://www.mdpi.com/2076-3417/15/10/5520 |
| work_keys_str_mv | AT borjaperez innovativeapproachestotrafficanomalydetectionandclassificationusingai AT marioresino innovativeapproachestotrafficanomalydetectionandclassificationusingai AT teresaseco innovativeapproachestotrafficanomalydetectionandclassificationusingai AT fernandogarcia innovativeapproachestotrafficanomalydetectionandclassificationusingai AT abdullaalkaff innovativeapproachestotrafficanomalydetectionandclassificationusingai |