Abstract:In the dynamic urban landscape, where the interplay of vehicles and pedestrians defines the rhythm of life, integrating advanced technology for safety and efficiency is increasingly crucial. This study delves into the application of cutting-edge technological methods in smart cities, focusing on enhancing public safety through improved traffic accident detection. Action recognition plays a pivotal role in interpreting visual data and tracking object motion such as human pose estimation in video sequences. The challenges of action recognition include variability in rapid actions, limited dataset, and environmental factors such as (Weather, Illumination, and Occlusions). In this paper, we present a novel comprehensive dataset for traffic accident detection. This datasets is specifically designed to bolster computer vision and action recognition systems in predicting and detecting road traffic accidents. We integrated datasets from wide variety of data sources, road networks, weather conditions, and regions across the globe. This approach is underpinned by empirical studies, aiming to contribute to the discourse on how technology can enhance the quality of life in densely populated areas. This research aims to bridge existing research gaps by introducing benchmark datasets that leverage state-of-the-art algorithms tailored for traffic accident detection in smart cities. These dataset is expected to advance academic research and also enhance real-time accident detection applications, contributing significantly to the evolution of smart urban environments. Our study marks a pivotal step towards safer, more efficient smart cities, harnessing the power of AI and machine learning to transform urban living.

What problem does this paper attempt to address?

This paper focuses on improving traffic safety, especially in terms of traffic accident detection, in smart cities through big data and deep learning. The research team has created a comprehensive dataset to enhance the capabilities of computer vision and action recognition systems in predicting and detecting road traffic accidents. This dataset combines various data sources, different weather conditions, and road networks from multiple regions worldwide to address the challenges posed by the variability of rapid actions, limited existing datasets, and environmental factors such as weather, lighting, and occlusion. The paper points out the limitations and research gaps in the field of action recognition, particularly in the context of traffic accident detection. Therefore, they propose a new dataset specifically designed for traffic accident detection, covering various traffic scenarios and accident types, including rear-end collisions, front-end collisions, side collisions, etc. Additionally, it includes videos of normal traffic conditions to enable the model to learn to differentiate between normal and abnormal situations. The paper also introduces existing traffic accident datasets and compares their characteristics, emphasizing the need for diverse and extensive data in order to improve model performance. The research team has meticulously annotated each video segment using the Labelbox tool to support the training and application of machine learning models, such as accident detection, autonomous driving, and traffic management systems. Finally, the paper presents a model called I3D-CONVLSTM2D, which utilizes the created dataset, showcasing how the integration of RGB frames and optical flow information can enhance the accuracy and efficiency of the accident detection system. The release of this dataset aims to facilitate the development of academic research and practical applications, contributing to the construction of safer and more efficient smart city environments.

Big Data and Deep Learning in Smart Cities: A Comprehensive Dataset for AI-Driven Traffic Accident Detection and Computer Vision Systems

Smart City Transportation: Deep Learning Ensemble Approach for Traffic Accident Detection

AI on the Road: A Comprehensive Analysis of Traffic Accidents and Accident Detection System in Smart Cities

A system of vision sensor based deep neural networks for complex driving scene analysis in support of crash risk assessment and prevention

VISION-BASED ACCIDENT IDENTIFICATION IN TRAFFIC VIDEOS USING DEEP LEARNING

Harnessing sensor fusion and AI for accurate accident detection and classification in the safety of smart cities

Urban Safety: An Image-Processing and Deep-Learning-Based Intelligent Traffic Management and Control System

TAD: A Large-Scale Benchmark for Traffic Accidents Detection from Video Surveillance

Vision-Based Accident Anticipation and Detection Using Deep Learning

A Real-Time Autonomous Highway Accident Detection Model Based on Big Data Processing and Computational Intelligence

Accident Risk Prediction based on Heterogeneous Sparse Data: New Dataset and Insights

Road Feature Detection for Advance Driver Assistance System Using Deep Learning

Harnessing Smartphone Sensors for Enhanced Road Safety: A Comprehensive Dataset and Review

Deep Learning-based Road Object Detection for Collision Avoidance in Autonomous Driving

Learning spatial patterns and temporal dependencies for traffic accident severity prediction: A deep learning approach

Applying Deep Learning to Detect Traffic Accidents in Real Time Using Spatiotemporal Sequential Data

Prevent Car Accidents by Using AI

Towards Big Data Analytics and Mining for UK Traffic Accident Analysis, Visualization & Prediction.

Predicting Crash Injury Severity in Smart Cities: a Novel Computational Approach with Wide and Deep Learning Model

TADS: a Novel Dataset for Road Traffic Accident Detection from a Surveillance Perspective

A Deep Learning Approach to the Citywide Traffic Accident Risk Prediction