Search Results (310)

Scene understanding is essential for enhancing driver safety, generating human-centric explanations for Automated Vehicle (AV) decisions, and leveraging Artificial Intelligence (AI) for retrospective driving video analysis. This study developed a dynamic scene retrieval system using Contrastive Language–Image Pretraining (CLIP) models, which can be optimized for real-time deployment on edge devices. The proposed system outperforms state-of-the-art in-context learning methods, including the zero-shot capabilities of GPT-4o, particularly in complex scenarios. By conducting frame-level analyses on the Honda Scenes Dataset, which contains a collection of about 80 h of annotated driving videos capturing diverse real-world road and weather conditions, our study highlights the robustness of CLIP models in learning visual concepts from natural language supervision. The results also showed that fine-tuning the CLIP models, such as ViT-L/14 (Vision Transformer) and ViT-B/32, significantly improved scene classification, achieving a top F1-score of 91.1%. These results demonstrate the ability of the system to deliver rapid and precise scene recognition, which can be used to meet the critical requirements of advanced driver assistance systems (ADASs). This study shows the potential of CLIP models to provide scalable and efficient frameworks for dynamic scene understanding and classification. Furthermore, this work lays the groundwork for advanced autonomous vehicle technologies by fostering a deeper understanding of driver behavior, road conditions, and safety-critical scenarios, marking a significant step toward smarter, safer, and more context-aware autonomous driving systems. Full article

In the current dynamic business landscape, digital transformation is recognized as a critical driver of entrepreneurship, innovation, and growth, particularly among small and medium-sized enterprises (SMEs). This study aims to investigate the key factors influencing digital transformation, focusing on their relevance in shaping strategic decisions and fostering innovation. Using a robust methodological approach, data were collected through an online survey, with Likert-scale questions assessing multiple dimensions of digital maturity across companies in various sectors of the Greek economy. The survey, conducted in the first semester of 2024, involved 156 companies from sectors such as retail, communication, technology, and public services, with significant representation from established organizations employing over 250 individuals and reporting annual turnovers exceeding EUR 50 million. The questionnaire items, adapted from existing validated scales, captured aspects such as digital skills, management intensity, business processes, innovation performance, departmental agility, and digital vision. By analyzing the responses, this study identifies critical drivers of digital transformation and highlights their role in guiding strategic decisions, emphasizing the evolving nature of digital entrepreneurship. The findings contribute to the broader discourse on digital transformation, offering actionable insights for organizations aiming to enhance their digital maturity and competitiveness in a rapidly changing global economy. Full article

This research investigates the role residents’ place attachment plays in developing their trust in the government in the city of Budapest. Rooted in different domains such as politics, environmental, and social psychology and aligned with the related SDGs (goals 3, 11 16, 17), an integrative model of residents’ place attachment, trust in the local government, QoL, social impacts, and pro-social engagement is proposed, which the authors argue are important mechanisms to be explored in determining support for sustainable tourism development to promote responsible consumption and production (SDG 12). We applied SEM on a sample of 350 residents. Findings suggest that place attachment significantly influences residents’ trust in the local government, which also has a positive influence on quality of life. Quality of life has a significant influence on perceived social impacts, which strongly leads to pro-social engagement; pro-social engagement underpins support for tourism development. This paper sets a future research agenda for sustainable tourism by indicating its possible antecedents. Sustainable tourism development has important implications for the achievement of other sustainability goals. The study aligns well with the United Nations Tourism’s vision of tourism as an important driver of positive change for responsible consumption and advancing other SDGs. Full article

Roadside incidents are a leading cause of driver fatalities in the United States, with a significant number involving collisions with barriers, such as guardrails. Guardrails are essential safety barriers designed to maintain vehicle trajectories and shield against roadside hazards. The functionality of guardrails heavily relies on their structural integrity, and damaged guardrails can pose serious dangers to road users. Traditional inspection methods are labor-intensive, time-consuming, and prone to human error, lacking periodic monitoring crucial for timely maintenance. Although advancements in computer vision have enabled automated infrastructure inspections, research dedicated specifically to the inspection of guardrails remains scarce. Existing automated solutions do not fully address the challenges of accurately identifying and assessing guardrail damage under varying lighting and weather conditions and the computational demands of real-time processing. This study addresses these challenges by introducing a novel framework utilizing advanced computer vision techniques, such as YOLOv8 models and the Deep OC–SORT tracker, integrated with camera and GPS systems mounted on a vehicle. This system automates the detection, localization, and severity assessment of guardrail damage, enhancing inspection accuracy and efficiency, enabling faster maintenance responses, and ultimately contributing to safer road conditions. Full article

Timely prediction of collisions enables advanced driver assistance systems to issue warnings and initiate emergency maneuvers as needed to avoid collisions. With recent developments in computer vision and deep learning, collision warning systems that use vision as the only sensory input have emerged. They are less expensive than those that use multiple sensors, but their effectiveness must be thoroughly assessed. We systematically searched academic literature for studies proposing ego-centric, vision-based collision warning systems that use deep learning techniques. Thirty-one studies among the search results satisfied our inclusion criteria. Risk of bias was assessed with PROBAST. We reviewed the selected studies and answer three primary questions: What are the (1) deep learning techniques used and how are they used? (2) datasets and experiments used to evaluate? (3) results achieved? We identified two main categories of methods: Those that use deep learning models to directly predict the probability of a future collision from input video, and those that use deep learning models at one or more stages of a pipeline to compute a threat metric before predicting collisions. More importantly, we show that the experimental evaluation of most systems is inadequate due to either not performing quantitative experiments or various biases present in the datasets used. Lack of suitable datasets is a major challenge to the evaluation of these systems and we suggest future work to address this issue. Full article

Due to the increase in the number of vehicles and the complexity of parking spaces, parking space detection technology has emerged. It is capable of automatically identifying vacant parking spaces in parking lots or on streets, and delivering this information to drivers or parking management systems in real time, which has a significant impact on improving urban parking efficiency, alleviating traffic congestion, optimizing driving experience, and promoting the development of intelligent transportation systems. This paper firstly describes the research significance of parking space detection technology and its research background, and then systematically reviews different types of parking spaces and detection technologies, covering a variety of technical means such as ultrasonic sensors, infrared sensors, magnetic sensors, other sensors, methods based on traditional computer vision, and methods based on deep learning. At the end of the paper, the article summarizes the current research progress in parking space detection technology, analyzes the existing challenges, and provides an outlook on future research directions. Full article

The use of traffic simulation to analyze traffic safety and performance has become common in transportation engineering. Microsimulation methods are increasingly used to analyze driving performance for different road geometries and environmental elements. Drivers’ perception has an important impact on driving performance factors contributing to traffic safety on transportation facilities (highways, arterials, intersections, etc.). Impaired vision leads to failure in drivers’ perception and making right decisions. Various studies investigated the impact of environmental elements (fog, rain, snow, etc.) on driving performance. However, there is limited research examining the potentially detrimental effects on driving capabilities due to differing exposure to natural light brightness, in particular sun exposure. Autonomous vehicles (AVs) showed a significant impact enhancing traffic capacity and improving safety margins in car-following models. AVs may also enhance and/or complement human driving under deteriorated driving conditions such as sun glare. This study uses a calibrated traffic simulation and surrogate safety assessment model to improve traffic operations and safety performance under impaired visibility using different types of autonomous vehicles. A combination of visibility reduction, traffic flow characteristics, and autonomy levels of AVs was simulated and assessed in terms of the number of conflicts, severity level, and traffic operations. The simulation analysis results used to reveal the contribution of conflicts to the risk of crashes varied based on the influence of autonomy level on safe driving during sun glare exposure. The outcome of this study indicates the benefits of using different levels of AVs as a solution to driving under vision impairment situations that researchers, traffic engineers, and policy makers can use to enhance traffic operation and road safety in urban areas. Full article

Saudi Vision 2030, a strategic framework aimed at diversifying the economy and enhancing societal inclusivity, aligns with the UN’s Sustainable Development Goals (SDGs) by promoting gender equality and sustainable economic growth. Sustainability is central to fostering women’s entrepreneurship, as it drives social equity, economic diversification, and innovation, elements which are crucial to sustainable development. While the existing literature has primarily focused on women’s entrepreneurship in the Western world, limited attention has been given to its development in the Global South, particularly in Saudi Arabia. As a nation undergoing transformative social, cultural, and economic shifts, women entrepreneurs play a critical role in aligning entrepreneurial efforts with global sustainability goals. This research investigates the factors influencing Saudi women to become entrepreneurs, specifically examining the factors that inspire or hinder them from creating their own ventures. Drawing upon cognitive and social capital theories, which have proven their soundness in the existing literature, this research utilizes a dataset of 1715 women entrepreneurs analyzed through binomial logistic regression. The findings indicate that social desirability, relational capital, experience as angel investors, age, income, and education significantly increase the likelihood of women’s entrepreneurship. By contextualizing women’s entrepreneurship within Saudi Arabia’s evolving societal and economic landscape, this research highlights their potential as drivers of inclusive growth and sustainable economic empowerment. Furthermore, the research outlines strategies to enhance women’s entrepreneurial participation, contributing both to the entrepreneurship literature and the realization of Saudi Vision 2030. Full article

Drowsy driving poses a significant challenge to road safety worldwide, contributing to thousands of accidents and fatalities annually. Despite advancements in driver drowsiness detection (DDD) systems, many existing methods face limitations such as intrusiveness and delayed reaction times. This research addresses these gaps by leveraging facial analysis and state-of-the-art machine learning techniques to develop a real-time, non-intrusive DDD system. A distinctive aspect of this research is its systematic assessment of various machine and deep learning algorithms across three pivotal public datasets, the NTHUDDD, YawDD, and UTA-RLDD, known for their widespread use in drowsiness detection studies. Our evaluation covered techniques including the K-Nearest Neighbors (KNNs), support vector machines (SVMs), convolutional neural networks (CNNs), and advanced computer vision (CV) models such as YOLOv5, YOLOv8, and Faster R-CNN. Notably, the KNNs classifier reported the highest accuracy of 98.89%, a precision of 99.27%, and an F1 score of 98.86% on the UTA-RLDD. Among the CV methods, YOLOv5 and YOLOv8 demonstrated exceptional performance, achieving 100% precision and recall with mAP@0.5 values of 99.5% on the UTA-RLDD. In contrast, Faster R-CNN showed an accuracy of 81.0% and a precision of 63.4% on the same dataset. These results demonstrate the potential of our system to significantly enhance road safety by providing proactive alerts in real time. Full article

Overtaking relies heavily on the driver’s attention and cognitive state, and illegal overtaking can lead to accidents, severe injuries, or fatalities. To enhance highway safety, we propose a method for accurately detecting illegal overtaking on continuous road lanes. We used dashboard-mounted smartphone cameras and geolocation data to filter the analysis areas. We used the state-of-the-art deep learning model You Only Look Once version 8 (YOLOv8) to detect yellow road lanes. When these lanes suggest potential illegal overtaking, we apply the YOLO for Panoptic driving Perception version 2 (YOLOPv2) model, followed by post-processing. We confirm overtaking events by checking for overlaps between detections from both models. We store confirmed instances and evaluate the information temporally rather than just from individual frames. We then analyze the entire video to identify violations and extract the moments of occurrence. We tested the algorithm on real-world traffic data under various weather and lighting conditions. Our method demonstrates reliability and consistency in identifying illegal overtaking. We achieved 16 TP and only 1 FP over 56 videos totaling 41 h, 9 min, and 24 s, with precision, recall, and F1-score values of 1.000, 0.941, and 0.970, respectively. Consequently, our innovative and practical solution, utilizing simple cameras and advanced computer vision models, can significantly enhance highway safety and support vehicle auditing systems. Full article

Economic, social, and environmental sustainability emphasizes the need for organizations to integrate sustainability strategies into their core business and business development plans. The era of Society 5.0 is characterized by human-centeredness and digital leadership. It requires embedding sustainability practices and Green Information Technology (IT) while leveraging human–technology relationships to promote social good. However, embedding these practices into organizational culture is challenging due to resistance to change and the need for widespread mindset shifts. This study selected a focus group of eight South African participants to define sociotechnical knowledge management (KM) guidelines for embedding sustainable practices in organizations to promote digital decarbonization aligned with the Society 5.0 vision. Our findings suggest ten elements for the guidelines to incorporate: external environment, organizational context, business drivers, business outcomes, monitoring and evaluation, KM processes, technology enablers, sociotechnical KM tactics, knowledge assets, and execution considerations. By adopting such guidelines as a sustainability strategy, organizations can integrate KM practices into the human-centered and cyber-physical philosophy of Society 5.0. This approach aligns employee behavior with technological tools, enabling organizations to make data-driven decisions, reduce digital waste, and foster a culture of environmental responsibility. In addition, this approach enhances collaboration and innovation, benefiting all stakeholders and advancing sustainable development. Full article

The significant number of road traffic accidents caused by fatigued drivers presents substantial risks to the public’s overall safety. In recent years, there has been a notable convergence of intelligent cameras and artificial intelligence (AI), leading to significant advancements in identifying driver drowsiness. Advances in computer vision technology allow for the identification of driver drowsiness by monitoring facial expressions such as yawning, eye movements, and head movements. These physical indications, together with assessments of the driver’s physiological condition and behavior, aid in assessing fatigue and lowering the likelihood of drowsy driving-related incidents. This study presents an extensive variety of meticulously designed algorithms that were thoroughly analyzed to assess their effectiveness in detecting drowsiness. At the core of this attempt lay the essential concept of feature extraction, an efficient technique for isolating facial and ocular regions from a particular set of input images. Following this, various deep learning models, such as a traditional CNN, VGG16, and MobileNet, facilitated detecting drowsiness. Among these approaches, the MobileNet model was a valuable choice for drowsiness detection in drivers due to its real-time processing capability and suitability for deployment in resource-constrained environments, with the highest achieved accuracy of 92.75%. Full article

Drowsy driving contributes to over 6000 fatal incidents annually in the US, underscoring the need for effective, non-intrusive drowsiness detection. This study seeks to address detection challenges, particularly in non-standard head positions. Our innovative approach leverages computer vision by combining facial feature detection using Dlib, head pose estimation with the HOPEnet model, and analyses of the percentage of eyelid closure over time (PERCLOS) and the percentage of mouth opening over time (POM). These are integrated with traditional machine learning models, such as Support Vector Machines, Random Forests, and XGBoost. These models were chosen for their ability to process detailed information from facial landmarks, head poses, PERCLOS, and POM. They achieved a high overall accuracy of 86.848% in detecting drowsiness, with a small overall model size of 5.05 MB and increased computational efficiency. The models were trained on the National Tsing Hua University Driver Drowsiness Detection Dataset, making them highly suitable for devices with a limited computational capacity. Compared to the baseline model from the literature, which achieved an accuracy of 84.82% and a larger overall model size of 37.82 MB, the method proposed in this research shows a notable improvement in the efficiency of the model with relatively similar accuracy. These findings provide a framework for future studies, potentially improving sleepiness detection systems and ultimately saving lives by enhancing road safety. Full article

The green vision rate of rural highway greening landscape is a key factor affecting the driver’s visual load. Based on this, this paper uses the eye tracking method to study the visual characteristics of drivers in different green vision environments on rural highways in Xianning County. Based on the HSV color space model, this paper obtains four sections of rural highway with a green vision rate of 10~20%, green vision rate of 20~30%, green vision rate of 30~40%, and green vision rate of 40~50%. Through the real car test, the pupil area, fixation time, saccade time, saccade angle, saccade speed, and other visual indicators of the driver’s green vision rate in each section were obtained. The visual load quantization model was combined with factor analysis to explore the influence degree of the green vision rate in each section on the driver’s visual load. The results show that the visual load of the driver in the four segments with different green vision rate is as follows: Z10~20% > Z20~30% > Z30~40% > Z40~50%. When the green vision rate is 10~20%, the driver’s fixation time becomes longer, the pupil area becomes larger, the visual load is the highest, and the driving is unstable. When the green vision rate is 40% to 50%, the driver’s fixation time and pupil area reach the minimum, the visual load is the lowest, and the driving stability is the highest. The research results can provide theoretical support for the design of rural highway landscape green vision rate and help to promote the theoretical research of traffic safety. Full article

Over the past few years, minor centers have attracted interest from the scientific community and beyond as places to be re-inhabited. They have started being regarded as places of healthy and wholesome living, places that have kept resilience to anthropic actions as well as a sensitive architectural and landscape heritage that can act as a driver for the socioeconomic regeneration of their territories if enhanced. Several initiatives network small neighboring municipalities and link them to various types of tourism (cultural, mountain, experiential, etc.), depending on the areas’ traditions and specific characteristics. However, minor centers are often still unprepared to welcome tourists and struggle to implement services, especially due to the economic deficit resulting from years of abandonment and depopulation. The research described here returns possible expeditious solutions for improving the condition of tourism-related services. Starting from the historical and urban analysis and knowledge acquisition of a specific case study—the Municipality of Naro, in Sicily—we reflected on solutions to be repeated in similar contexts to improve the accessibility and use of the historic center. The aim of the research is to outline a place-based design to improve mobility, water and waste management services, affecting places’ attractiveness. The proposed interventions are modular, increasable in small steps, with budgets suited to the economic possibilities of small centers such as the examined one. This method, due to its incremental and adaptive nature, is working ‘on’ places and ‘for’ places, as well as functions as a possible tool and stimulus for the self-construction of a ‘sustainable society’ that helps the governance of these centers toward a vision of urban valorization. Territories like Naro represent a large part of Italy. They are endowed with resources lacking in heavily urbanized areas yet involved in numerous revitalization policies, including international ones. Full article