Road Hazards Research Articles

Metaheuristic optimized complex-valued dilated recurrent neural network for attack detection in internet of vehicular communications

The Internet of Vehicles (IoV) is a specialized iteration of the Internet of Things (IoT) tailored to facilitate communication and connectivity among vehicles and their environment. It harnesses the power of advanced technologies such as cloud computing, wireless communication, and data analytics to seamlessly exchange real-time data among vehicles, road-side infrastructure, traffic management systems, and other entities. The primary objectives of this real-time data exchange include enhancing road safety, reducing traffic congestion, boosting traffic flow efficiency, and enriching the driving experience. Through the IoV, vehicles can share information about traffic conditions, weather forecasts, road hazards, and other relevant data, fostering smarter, safer, and more efficient transportation networks. Developing, implementing and maintaining sophisticated techniques for detecting attacks present significant challenges and costs, which might limit their deployment, especially in smaller settings or those with constrained resources. To overcome these drawbacks, this article outlines developing an innovative attack detection model for the IoV using advanced deep learning techniques. The model aims to enhance security in vehicular networks by efficiently identifying attacks. Initially, data is collected from online databases and subjected to an optimal feature extraction process. During this phase, the Enhanced Exploitation in Hybrid Leader-based Optimization (EEHLO) method is employed to select the optimal features. These features are utilized by a Complex-Valued Dilated Recurrent Neural Network (CV-DRNN) to detect attacks within vehicle networks accurately. The performance of this novel attack detection model is rigorously evaluated and compared with that of traditional models using a variety of metrics.

Automated Recognition of Snow-Covered and Icy Road Surfaces Based on T-Net of Mount Tianshan

The Tianshan Expressway plays a crucial role in China’s “Belt and Road” strategy, yet the extreme climate of the Tianshan Mountains poses significant traffic safety risks, hindering local economic development. Efficient detection of hazardous road surface conditions (RSCs) is vital to address these challenges. The complexity and variability of RSCs in the region, exacerbated by harsh weather, make traditional surveillance methods inadequate for real-time monitoring. To overcome these limitations, a vision-based artificial intelligence approach is urgently needed to ensure effective, real-time detection of dangerous RSCs in the Tianshan road network. This paper analyzes the primary structures and architectures of mainstream neural networks and explores their performance for RSC recognition through a comprehensive set of experiments, filling a research gap. Additionally, T-Net, specifically designed for the Tianshan Expressway engineering project, is built upon the optimal architecture identified in this study. Leveraging the split-transform-merge structure paradigm and asymmetric convolution, the model excels in capturing detailed information by learning features across multiple dimensions and perspectives. Furthermore, the integration of channel, spatial, and multi-head attention modules enhances the weighting of key features, making the T-Net particularly effective in recognizing the characteristics of snow-covered and icy road surfaces. All models presented in this paper were trained on a custom RSC dataset, compiled from various sources. Experimental results indicate that the T-Net outperforms fourteen once state-of-the-art (SOTA) models and three models specifically designed for RSC recognition, with 97.44% accuracy and 9.79% loss on the validation set.

IMPROVING ROAD SAFETY AT ACCIDENT-PRONE AREAS: A COMPARISON BETWEEN GLOW-IN-THE-DARK AND CONVENTIONAL ROAD MARKING

Driving at night is challenging due to poor vision, poor road visual guidance, and the need to encounter bright light sources. Factors like lack of street lighting, fading, and lack of reflectivity by road studs and retro-reflective materials affect road users' vision at night. Commentators have pointed out there is a critical demand to improve road safety in preventing road accidents and hazards that caused by the poor visibility of road markings and inadequacy lighting. Although road safety management has been given an important focus by the Ministry of Transportation Malaysia to achieve 50% reduction of road accidents by 2030, there is a critical demand to improve the present road markings method to overcome the issues of poor visibility or unclear road markings and inadequacy lighting that may trigger potential hazards to road users at night, with a new innovative road marking technology. Therefore, this paper is prepared with the objective to present the outcome of comparative study between the present conventional road markings and the newly innovative technology of glow-in-the-dark method with specific reference to the road safety management in Malaysia. This study employed a questionnaire survey to interview fifty-one (51) respondents that have experience of driving at night, with the objectives to get their opinions on present condition of road markings in Malaysia, and how the newly innovative technology glow-in-the-dark can improve road safety in Malaysia. The study outcome revealed that the glow-in-the-dark is deemed appropriate to enhance the visibility during low-light conditions as compared to the present conventional road markings. Although the initial cost is higher, the economic advantages of glow-in-the-dark in reducing the streetlight usage, lower electrical costs, and substantially extended lifespan have made this newly innovative road marking outweighed the downsides and economically wise choice to revolutionize road safety management practice at the accident-prone areas in Malaysia for achieving the SDG 3: Good Health and Well-being.

Distracted Driving Behavior Detection Algorithm Based on Lightweight StarDL-YOLO

Distracted driving is one of the major factors leading drivers to ignore potential road hazards. In response to the challenges of high computational complexity, limited generalization capacity, and suboptimal detection accuracy in existing deep learning-based detection algorithms, this paper introduces a novel approach called StarDL-YOLO (StarNet-detectlscd-yolo), which leverages an enhanced version of YOLOv8n. Initially, the StarNet integrated into the backbone of YOLOv8n significantly improves the feature extraction capability of the model with remarkable reduction in computational complexity. Subsequently, the Star Block is incorporated into the neck network, forming a C2f-Star module that offers lower computational cost. Additionally, shared convolution is introduced in the detection head to further reduce computational burden and parameter size. Finally, the Wise-Focaler-MPDIoU loss function is proposed to strengthen detection accuracy. The experimental results demonstrate that StarDL-YOLO significantly improves the efficiency of the distracted driving behavior detection, achieving an accuracy of 99.6% on the StateFarm dataset. Moreover, the parameter count of the model is minimized by 56.4%, and its computational load is decreased by 45.1%. Additionally, generalization experiments are performed on the 100-Driver dataset, revealing that the proposed scheme enhances generalization effectiveness compared to YOLOv8n. Therefore, this algorithm significantly reduces computational load while maintaining high reliability and generalization capability.

Lane Detection, Segmentation, Pothole Detection and Traffic Sign Recognition for ADAS

Lane detection, a critical aspect of advanced driver assistance systems (ADAS) and autonomous vehicles, is addressed in this work, combining classical computer vision methods with deep learning techniques. The proposed solution, utilizing a UNet-based model trained with TensorFlow and Keras, enhances vehicle perception for intelligent transportation systems. Integrated functionalities for pothole detection and traffic sign detection further contribute to safety and efficiency, enabling vehicles to identify road hazards and comply with regulations. The system encompasses data preprocessing, model training, and real-time video analysis, while a classical lane detection pipeline using OpenCV showcases various stages such as grayscale conversion, Gaussian blur, Canny edge detection, masking, Hough transform, and lane overlay. This comprehensive approach supports road safety, traffic management, and transportation efficiency initiatives, making significant strides in intelligent transportation systems and urban planning.

Using Support Vector Machines to Classify Road Surface Conditions to Promote Safe Driving.

Accurate detection of road surface conditions in adverse winter weather is essential for traffic safety. To promote safe driving and efficient road management, this study presents an accurate and generalizable data-driven learning model for the estimation of road surface conditions. The machine model was a support vector machine (SVM), which has been successfully applied in diverse fields, and kernel functions (linear, Gaussian, second-order polynomial) with a soft margin classification technique were also adopted. Two learner designs (one-vs-one, one-vs-all) extended their application to multi-class classification. In addition to this non-probabilistic classifier, this study calculated the posterior probability of belonging to each group by applying the sigmoid function to the classification scores obtained by the trained SVM. The results indicate that the classification errors of all the classifiers, excluding the one-vs-all linear learners, were below 3%, thereby accurately classifying road surface conditions, and that the generalization performance of all the one-vs-one learners was within an error rate of 4%. The results also showed that the posterior probabilities can analyze certain atmospheric and road surface conditions that correspond to a high probability of hazardous road surface conditions. Therefore, this study demonstrates the potential of data-driven learning models in classifying road surface conditions accurately.

Road hazard: a systematic review of traffic injuries following recreational cannabis legalization.

Acute cannabis use is associated with impaired driving performance and increased risk of motor vehicle crashes. Following the Canadian Cannabis Act's implementation, it is essential to understand how recreational cannabis legalization impacts traffic injuries, with a particular emphasis on Canadian emergency departments. This study aims to assess the impact of recreational cannabis legalization on traffic-related emergency department visits and hospitalizations in the broader context of North America. A systematic review was conducted according to best practices and reported using PRISMA 2020 guidelines. The protocol was registered on July 5, 2022 (PROSPERO CRD42022342126). MEDLINE(R) ALL (OvidSP), Embase (OvidSP), CINAHL (EBSCOHost), and Scopus were searched without language or date restrictions up to October 12, 2023. Studies were included if they examined cannabis-related traffic-injury emergency department visits and hospitalizations before and after recreational cannabis legalization. The risk of bias was assessed. Meta-analysis was not possible due to heterogeneity. Seven studies were eligible for the analysis. All studies were conducted between 2019 and 2023 in Canada and the United States. We found mixed results regarding the impact of recreational cannabis legalization on emergency department visits for traffic injuries. Four of the studies included reported increases in traffic injuries after legalization, while the remaining three studies found no significant change. There was a moderate overall risk of bias among the studies included. This systematic review highlights the complexity of assessing the impact of recreational cannabis legalization on traffic injuries. Our findings show a varied impact on emergency department visits and hospitalizations across North America. This underlines the importance of Canadian emergency physicians staying informed about regional cannabis policies. Training on identifying and treating cannabis-related impairments should be incorporated into standard protocols to enhance response effectiveness and patient safety in light of evolving cannabis legislation.

An Assessment of Road Hazard and Challenges in Nigeria. A Case Study of LERE LOCAL Government of Kaduna State

An Assessment of Road Hazard and Challenges in Nigeria. A Case Study of LERE LOCAL Government of Kaduna State

Semi-Autonomous Smart Electric Vehicle with AI based Drive Assistance

In our research, we propose a practical methodology design and implementation of a semi-autonomous smart electric vehicle (EV) equipped with advanced features such as AI-based pothole detection, a robust battery management system (BMS), self-parking capability, and intelligent headlight intensity control. The primary objective of this research is to enhance the safety, efficiency, and user experience of electric vehicles through the integration of cutting-edge technologies.The AI-based pothole detection system utilizes sensors and machine learning algorithms to continuously monitor the road surface for hazards such as potholes, cracks, or debris. By analyzing real-time sensor data, the system can accurately identify potential road hazards and provide timely warnings to the driver or take proactive measures to mitigate their impact on vehicle performance and passenger comfort. The battery management system plays a critical role in optimizing the performance and longevity of the vehicle’s battery pack. Through advanced algorithms and predictive analytics, the BMS monitors various parameters such as temperature, voltage, and current to ensure safe and efficient operation of the battery. By dynamically adjusting charging rates and load distribution, the system maximizes the battery’s lifespan while ensuring sufficient energy availability for propulsion and onboard systems. Self-parking technology enables the vehicle to autonomously navigate parking spaces and execute precise maneuvers without human intervention.

Does using a bioptic telescope improve visual recognition of the on-road environment?

Individuals with reduced visual acuity (VA) may drive with bioptic telescopes in some jurisdictions. The effect of bioptic telescope use on on-road recognition distances is important for driving safety, as increased recognition distances increase the time available to react to road signs or driving events. This study aimed to investigate the impact of bioptic telescope use on visual recognition of road signs, traffic lights, and hazards in an on-road driving environment in individuals with reduced VA. Ten individuals (mean ± standard deviation age, 39.1 ± 19.1 years) with reduced VA (trained to use bioptic telescopes) participated in a repeated-measures on-road experiment for two viewing conditions, with and without a bioptic telescope (randomized order). Participants underwent a clinical low vision assessment, including high-contrast VA with and without the bioptic telescope aligned, visual fields, and contrast sensitivity testing. For the driving component, participants seated in the front passenger seat of a moving car reported all road signs, traffic lights, and hazards (unpredictable road events that represent a risk to driving safety including other vehicles, cyclists, pedestrians) seen along a route that included suburban roads and highways. Video cameras captured participants' viewing behavior, verbal commentary, and the distance at which three pre-selected road signs were reported. In the eye with the bioptic telescope, high-contrast VA improved from 0.75 ± 0.17 (without) to 0.25 ± 0.1 logMAR with the bioptic telescope. Bioptic telescope use did not affect the percentage of road signs, traffic lights, or hazards correctly recognized but did result in 2.6 times longer recognition distances (49 ± 23 vs. 19 ± 11 m, t9 = 5.02, p<0.001). Road signs were recognized at significantly longer distances when using a bioptic telescope, confirming their positive impact on timely visual recognition of objects within the driving scene. Future work should explore whether this effect generalizes to individuals using bioptic telescopes when driving a vehicle.

Geostatistical Spatial Decision-Making for Identifying Road Hazardous Road Segments in Rural Areas

Abstract Road safety has now become a major concern in many countries. In Algeria, as elsewhere, many factors are involved in the phenomena of road accidents. In this field of research, our study aims to spatial identification along a road network the hazardous road segments which present a dangerousness index. We proceeded the use of geostatistical approach, in particular global G(d) statistic and local Gi*(d) statistic of Getis-Ord. However, in an operational framework, this approach is highly dependent on hypothesis on the process of the measurement of the spatial proximity based on the distances between the BSUs. Thus, the interpretation of empirical results based on the optimal choice of the weighting structure and the critical distance. Methodologically, this work allowed to test an approach for assess the safety levels experimented in the case of road network specifically higher accidents risk in the Wilaya of Mascara (north west of Algeria). The first results confirm the validity of the method and allow a recoverable mapping by managers in the choice of measures of preventive management.

Advancing Road Safety through Cloud Based RSU Solutions for Smart Internet of Vehicles

This research proposes a system to improve road safety by detecting and alerting drivers about hazardous road conditions in real-time. Vehicles equipped with sensors, GPS, and communication devices can autonomously detect hazardous road conditions and transmit alerts to a central cloud system through IoT MQTT protocol. Upon receiving alerts, vehicles can dynamically adjust their routes to avoid hazardous areas, reducing the risk of accidents. By utilizing IoT technology, the system ensures the reliability and authenticity of shared information, enhancing trust and overall road safety awareness. The proposed system architecture leverages cloud-based Roadside Units (RSUs) to facilitate communication among smart vehicles, providing real-time road condition information and ensuring secure and efficient data exchange. The system aims to address the challenges of existing IoV frameworks, such as data protection, key management, storage, concurrency performance, and response speed, by implementing robust security measures and efficient data management strategies.

VIRTUAL POTHOLE DETECTION SYSTEM

The intersection of artificial intelligence and vehicular technology has birthed a realm of innovation, revolutionizing our engagement with the automotive sphere. This survey delves into the realm of virtual pothole detection systems, a pioneering endeavor aimed at mitigating road hazards through technological intervention. As society gravitates towards a digitalized landscape, the imperative for enhanced road safety looms large. Virtual pothole detection systems represent a concerted effort to address this pressing concern, leveraging cutting-edge advancements in sensor technology, data analytics, and vehicular communication protocols. This paper scrutinizes the evolution, methodologies, and efficacy of virtual pothole detection systems, offering insights into their operational paradigms and prospects. Amidst the backdrop of burgeoning urbanization and escalating vehicular traffic, the significance of such systems cannot be overstated. By elucidating the intricacies of virtual pothole detection systems, this survey endeavors to catalyze further research and innovation in the pursuit of safer, more resilient road networks Keywords: Virtual Pothole Detection System (VPDS), Sensor Technology, Data Analytics, Vehicular Communication, Road Safety, Pothole Detection, Accident Prevention, Road Maintenance, Artificial Intelligence.

Machine vision-based autonomous road hazard avoidance system for self-driving vehicles

The resolution of traffic congestion and personal safety issues holds paramount importance for human’s life. The ability of an autonomous driving system to navigate complex road conditions is crucial. Deep learning has greatly facilitated machine vision perception in autonomous driving. Aiming at the problem of small target detection in traditional YOLOv5s, this paper proposes an optimized target detection algorithm. The C3 module on the algorithm’s backbone is upgraded to the CBAMC3 module, introducing a novel GELU activation function and EfficiCIoU loss function, which accelerate convergence on position loss lbox, confidence loss lobj, and classification loss lcls, enhance image learning capabilities and address the issue of inaccurate detection of small targets by improving the algorithm. Testing with a vehicle-mounted camera on a predefined route effectively identifies road vehicles and analyzes depth position information. The avoidance model, combined with Pure Pursuit and MPC control algorithms, exhibits more stable variations in vehicle speed, front-wheel steering angle, lateral acceleration, etc., compared to the non-optimized version. The robustness of the driving system's visual avoidance functionality is enhanced, further ameliorating congestion issues and ensuring personal safety.

Identification of hazardous road sites: a comparison of blackspot methodology of Narogong Road Bekasi and Johor Federal Roads

Identification of hazardous road sites: a comparison of blackspot methodology of Narogong Road Bekasi and Johor Federal Roads

Module lattice based post quantum secure blockchain empowered vehicle to vehicle communication in the internet of vehicles

The importance of vehicle-to-vehicle (V2V) communication is increasing due to high rate of accidents and advancements in information exchange system. The V2V communication facilitates vehicles to exchange real-time informations about their surrounding, such as nearby vehicle, pedestrian, and road hazard. These type of informations are very helpful to detect potential collision, warn driver of imminent danger, and take advanced measure to avoid accident. Both security and privacy of user participating in the vehicular networking system depends upon reliable data transmission system. The suggested approaches must have minimal processing complexity and be unaffected by latency problems due to the scalability of V2V connection. This paper contains module lattice based quantum safe blockchain empowered authentication framework for autonomous vehicle communication system. This framework uses blockchain that is always contains multiple nodes, then we need an aggregation technique to provide authenticity to the system. The proposed framework provides authenticity to the system based on two assumptions, (1) Module Learning With Error and (2) Module Short Integer Solution. The proposed framework is quantum safe, provides high latency and throughput.

Automatic detection of potholes using VGG-16 pre-trained network and Convolutional Neural Network

A self-driving car is necessary to implement traffic intelligence because it can vastly enhance both the safety of driving and the comfort of the driver by adjusting to the circumstances of the road ahead. Road hazards such as potholes can be a big challenge for autonomous vehicles, increasing the risk of crashes and vehicle damage. Real-time identification of road potholes is required to solve this issue. To this end, various approaches have been tried, including notifying the appropriate authorities, utilizing vibration-based sensors, and engaging in three-dimensional laser imaging. Unfortunately, these approaches have several drawbacks, such as large initial expenditures and the possibility of being discovered. Transfer learning is considered a potential answer to the pressing necessity of automating the process of pothole identification. A Convolutional Neural Network (CNN) is constructed to categorize potholes effectively using the VGG-16 pre-trained model as a transfer learning model throughout the training process. A Super-Resolution Generative Adversarial Network (SRGAN) is suggested to enhance the image's overall quality. Experiments conducted with the suggested approach of classifying road potholes revealed a high accuracy rate of 97.3%, and its effectiveness was tested using various criteria. The developed transfer learning technique obtained the best accuracy rate compared to many other deep learning algorithms.

A Review on Highway Network System - Samruddhi Expressway

Highway networks serve as critical arteries of transportation infrastructure, facilitating the movement of goods and people across vast geographical regions. As traffic volumes continue to increase and technological advancements reshape transportation systems, there is a pressing need to optimize the efficiency, safety, and sustainability of highway networks. This abstract presents a comprehensive system approach aimed at addressing these challenges. The proposed highway network system integrates advanced technologies such as intelligent transportation systems (ITS), real-time traffic monitoring, data analytics, and machine learning algorithms to optimize traffic flow, minimize congestion, and enhance safety. By leveraging data from various sources including sensors, cameras, and connected vehicles, the system enables proactive traffic management and congestion prediction, allowing authorities to implement timely interventions and mitigate potential disruptions. Furthermore, the integration of artificial intelligence and machine learning algorithms enables predictive maintenance of infrastructure, identifying potential issues before they escalate into critical failures. This proactive approach not only reduces maintenance costs but also enhances the reliability and lifespan of highway assets. Moreover, the system prioritizes safety through the implementation of automated collision avoidance systems, adaptive speed control, and real-time incident management. By leveraging vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communication technologies, the system enables vehicles to exchange critical information, such as hazardous road conditions or impending collisions, thereby reducing the risk of accidents and improving overall safety. Additionally, the system incorporates sustainability principles by optimizing traffic flow patterns to minimize fuel consumption and emissions, promoting eco-friendly transportation practices, and facilitating the integration of electric and autonomous vehicles into the existing infrastructure.

POTHOLE DETECTION SYSTEM USING MACHINE LEARNING

Potholes represent a ubiquitous road hazard, posing risks to both vehicular safety and infrastructure integrity. Addressing this challenge requires efficient and automated detection systems. Leveraging advancements in machine learning (ML), this paper proposes a pothole detection system using ML techniques. The system employs feature extraction, convolutional neural networks, transfer learning, and semantic segmentation for robust and accurate detection of potholes from image data. The proposed system contributes to the automation of road maintenance processes, enabling timely repairs and enhancing road safety. Through extensive experimentation and evaluation, the system demonstrates promising results in terms of detection accuracy and efficiency. Furthermore, discussions on integration with smart city infrastructure and autonomous vehicles highlight its potential impact on enhancing transportation systems' resilience and safety. This paper provides insights into the development and deployment of ML-based pothole detection systems, paving the way for improved road maintenance strategies and safer travel experiences. Key Words: Feature extraction, Convolutional Neural Network, Transfer learning, Semantic segmentation

Privacy-preserving dual interactive wasserstein generative adversarial network for cloud-based road condition monitoring in VANETs

Nowadays, a numerous fatalities have resulted from road damage. Research into road damage detection, particularly the detection and notification of hazardous road damage is essential for enhancing the traffic security. Existing road damage identification schemes often process data in the cloud, but they are unable to warn users on time on account of lengthy latency. Although newer edge computing strategies reduce this issue owing to the limited communication range of edges, the users can only get notifications about potentially harmful road damage within a small radius. Besides, untrusted nodes may misuse the sensitive information of users. Therefore, in this manuscript, a Privacy-preserving Dual Interactive Wasserstein Generative Adversarial Network is proposed for Cloud-Based Road Condition Monitoring in VANETs (PP-DIWGAN-C-RCM-VANET). Here, local models and global models are the two types of cloud detection models are considered. In which, the local models acquire knowledge from cloud-based data, whereas global models support local models' learning by combining local models. The intention of the proposed approach is “to detect the smart hazardous road condition”. The local model recognizes dangerous roads, and also categorized into 3 levels: low, middle, and high, these are based on information gathered about the state of the roads. From the road crack dataset, it is categorized into crack, normal and pothole. DIWGAN secures secrecy from unreliable clouds, but it is still significant danger of privacy leaking from unreliable edges. To ensure privacy, users must erase all data before transmitting it to the edges, here the user indicates the owner of the vehicle. Therefore, a new privacy preserving method called Fractional Discrete Meixner Moments Encryption (FDMME) is taken into account to fulfill this requirement. For real time dataset, the proposed method achieves better accuracy 27.5%, 10.32% and 16.65%, better f-score 30.93%, 11.14% and 15.3% compared with existing methods.