Electric vehicles: Battery technologies, charging standards, AI communications, challenges, and future directions

Energy Conversion and Management X, Journal Year: 2024, Volume and Issue: 24, P. 100751 - 100751

Published: Oct. 1, 2024

Language: Английский

---

Novel Transfer Learning Approach for Driver Drowsiness Detection Using Eye Movement Behavior

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 64765 - 64778

Published: Jan. 1, 2024

Driver drowsiness detection is a critical field of research within automotive safety, aimed at identifying signs fatigue in drivers to prevent accidents. Drowsiness impairs driver's reaction time, decision-making ability, and overall alertness, significantly increasing the risk collisions. Nowadays, challenge detect using physiological signals, which often require direct contact with body. This can be uncomfortable distracting. study detecting driver through eye movement behavior imagery driver. We utilized standard image dataset based on conduct this experiment. proposed novel transfer learning-based features generation combined strengths Visual Geometry Group (VGG-16) Light Gradient-Boosting Machine (LGBM) methods. The VGLG (VGG16-LGBM) approach first extracts spatial from input data then generates salient LGBM. Experimental evaluations reveal that k-neighbors classifier outperformed state-of-the-art high-performance accuracy 99%. computational complexity analysis shows detects 0.00829 seconds. have enhanced performance hyperparameter tuning validations k-fold validation. has potential revolutionize detection, aiming road accidents save precious lives.

Language: Английский

---

Processing and Integration of Multimodal Image Data Supporting the Detection of Behaviors Related to Reduced Concentration Level of Motor Vehicle Users

Electronics, Journal Year: 2024, Volume and Issue: 13(13), P. 2457 - 2457

Published: June 23, 2024

This paper introduces a comprehensive framework for the detection of behaviors indicative reduced concentration levels among motor vehicle operators, leveraging multimodal image data. By integrating dedicated deep learning models, our approach systematically analyzes RGB images, depth maps, and thermal imagery to identify driver drowsiness distraction signs. Our novel contribution includes utilizing state-of-the-art convolutional neural networks (CNNs) bidirectional long short-term memory (Bi-LSTM) effective feature extraction classification across diverse scenarios. Additionally, we explore various data fusion techniques, demonstrating their impact on improving accuracy. The significance this work lies in its potential enhance road safety by providing more reliable efficient tools real-time monitoring attentiveness, thereby reducing risk accidents caused fatigue. proposed methods are thoroughly evaluated using benchmark dataset, with results showing substantial capabilities leading development safety-enhancing technologies vehicular environments. primary challenge addressed study is states not relying lighting conditions. solution employs integration, encompassing RGB, thermal, ensure robust accurate regardless external variations

Language: Английский

---

A review of deep learning in blink detection

PeerJ Computer Science, Journal Year: 2025, Volume and Issue: 11, P. e2594 - e2594

Published: Jan. 14, 2025

Blink detection is a highly concerned research direction in the field of computer vision, which plays key role various application scenes such as human-computer interaction, fatigue and emotion perception. In recent years, with rapid development deep learning, learning techniques for precise blink has emerged significant area interest among researchers. Compared traditional methods, method based on offers superior feature ability higher accuracy. However, current lacks systematic summarization comparison. Therefore, aim this article to comprehensively review progress learning-based methods help researchers have clear understanding approaches field. This analyzes made by several classical models practical applications eye while highlighting their respective strengths weaknesses. Furthermore, it provides comprehensive summary commonly used datasets evaluation metrics detection. Finally, discusses challenges future directions applications. Our analysis reveals that demonstrate strong performance they encounter challenges, including training data imbalance, complex environment interference, real-time processing issues device limitations. By overcoming identified study, prospects algorithms will be significantly enhanced.

Language: Английский

---

Real-Time Driver Drowsiness Detection Using Facial Analysis and Machine Learning Techniques

Sensors, Journal Year: 2025, Volume and Issue: 25(3), P. 812 - 812

Published: Jan. 29, 2025

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

Language: Английский

---

The Role of Deep Learning Models in the Detection of Anti-Social Behaviours towards Women in Public Transport from Surveillance Videos: A Scoping Review

Safety, Journal Year: 2023, Volume and Issue: 9(4), P. 91 - 91

Published: Dec. 13, 2023

Increasing women’s active participation in economic, educational, and social spheres requires ensuring safe public transport environments. This study investigates the potential of machine learning-based models addressing behaviours impacting safety perception women commuters. Specifically, we conduct a comprehensive review existing literature concerning utilisation deep learning for identifying anti-social spaces. Employing scoping methodology, our synthesises current landscape, highlighting both advantages challenges associated with automated detection such behaviours. Additionally, assess available video audio datasets suitable training algorithms this context. The findings not only shed light on feasibility leveraging recognising but also provide critical insights researchers, developers, operators. Our work aims to facilitate future studies focused development implementation models, enhancing all passengers transportation systems.

Language: Английский

---

COOT-Optimized Real-Time Drowsiness Detection using GRU and Enhanced Deep Belief Networks for Advanced Driver Safety

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(4)

Published: Jan. 1, 2024

Drowsiness among drivers is a major hazard to road safety, resulting in innumerable incidents globally. Despite substantial study, existing approaches for detecting drowsiness real time continue confront obstacles, such as low accuracy and efficiency. In these circumstances, this study tackles the critical problems of identifying driver safety by suggesting novel approach that leverages combined effectiveness Gated Recurrent Units (GRU) Enhanced Deep Belief Networks (EDBN), which optimised using COOT, new bird collective-behavioral-based optimisation algorithm. The begins emphasising relevance sleepiness detection improving limitations prior studies reaching high real-time detection. suggested method tries close gap combining GRU EDBN simulations, are known their temporal modelling feature learning capabilities, respectively, give comprehensive solution Following thorough experimentation, technique achieves an outstanding around 99%, indicating its efficiency states driving scenarios. research stems from potential greatly reduce number accidents caused drowsy driving, hence overall safety. Furthermore, use COOT optimize parameters models adds dimension research, demonstrating nature-inspired optimization methodologies performance machine algorithms applications

Language: Английский

---

Drowsiness and Emotion Detection of Drivers for Improved Road Safety

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 13 - 26

Published: Jan. 1, 2024

Language: Английский

---

A Convolutional Neural Network-Based Method for Real- Time Eye State Identification in Driver Drowsiness Detection

Published: Dec. 29, 2023

The use of Convolutional Neural Networks (CNNs) for drowsiness detection is an AI-based method detecting and warning human or weariness. Deep learning models like convolutional neural networks learn by analysing images videos, often focusing on faces eyes. Drowsiness cues may be recognised these models, including drooping eyelids, yawning, altered facial expressions. CNN-based sleepiness used to generate alarms cautions using economic resources in real-time applications, such as driver monitoring systems, help avoid accidents improve safety. Artificial intelligence computer vision access are at the heart this technology, which aims alleviate major safety risks associated with exhaustion. In research, classification accomplished a Sequential Network (CNN) model. graphics processing unit (GPU) producing preliminary data processing. Learning model then shown graphically Loss Accuracy curves. With predicted accuracy 96%, suggested would pave way more research into categorization.

Language: Английский

---

Early Detection of Diabetic Retinopathy Utilizing Advanced Fuzzy Logic Techniques

Mathematical Modelling and Engineering Problems, Journal Year: 2023, Volume and Issue: 10(6), P. 2086 - 2094

Published: Dec. 21, 2023

The escalating prevalence of diabetes globally, exacerbated by lifestyle changes postpandemic-including increased screen time, sedentary behavior, and remote workhas consequently driven a surge in associated complications, notably, Diabetic Retinopathy (DR).This ocular complication presents pressing concern due to its potential precipitate irreversible vision loss.Consequently, the necessity for timely accurate DR detection is paramount, especially circumstances where conventional diagnostic approaches are either challenging or financially prohibitive.Capitalizing on prowess fuzzy logic managing uncertainties, this study introduces an innovative application Extended Fuzzy Logic early-stage DR.Rather than focusing solely overt symptoms, approach discerns subtle similarities retinal irregularities between diabetic patients non-diabetic individuals.To quantify these similarities, 'f-validity' value was computed based risk factors which were subsequently transformed into membership function values.The aggregation values facilitated Ordered Weighted Averaging (OWA) operator.The experimental outcomes align satisfactorily with expert anticipations, boasting accuracy 90%, precision 92.2%, sensitivity 75%.These results, when juxtaposed against contemporary studies field, underscore promise scheme advancing early diagnostics DR.The thus proposes solution that leverages power address burgeoning challenge DR.

Language: Английский

---