Segun Adebayo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Segun Adebayo has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Halleluyah Oluwatobi Aworinde
Department of Information Technology, Durban University of Technology, Durban 4000, South Africa
9 shared publications - Abidemi Emmanuel Adeniyi
Department of Information Systems, Faculty of Computer Science and Engineering, Obafemi Awolowo University, Ile-Ife 220101, Nigeria
5 shared publications Aderonke B. Sakpere
University of Ibadan
3 shared publications- Akinwale O. Akinwunmi
College of Computing and Communication Studies, Bowen University, Iwo Nigeria
3 shared publications - Bukola Oyerinola Atobatele
Microbiology, Department, College of Agriculture, Engineering and Science, Bowen University, Iwo, Nigeria
3 shared publications - Oluwasegun Julius Aroba
Durban University of Technology
3 shared publications
Latest Publications
- 2 August 2026
- Published by Springer Nature
- p. 248-261
Development of a Smart Classroom Monitoring System for Responsive and Adaptive Learning Using TinyML
- 2 August 2026
- Published by Springer Nature
- p. 248-261
Development of a Smart Classroom Monitoring System for Responsive and Adaptive Learning Using TinyML
- Open Access
- 28 July 2026
- Published by Elsevier in SSRN Electronic Journal
A Sustainability-Oriented Predictive Maintenance Framework for Renewable Energy Actuation Systems
Background In renewable energy systems such as wind turbines, solar tracking systems, and energy storage facilities, electromechanical and electro-pneumatic actuation subsystems play a crucial role. These actuator failures lead to lower energy production efficiency, higher maintenance costs, and higher carbon emissions because of unexpected downtime and frequent maintenance. Due to this, there is an increasing demand for intelligent predictive maintenance strategies that can spot incipient faults before they become critical failures. Methods In this study, a framework for predictive maintenance through model-based dynamic analysis and Support Vector Machine (SVM) based incipient fault detection for electro-pneumatic actuation systems in renewable energy applications is presented. The actuator dynamics of a double-acting cylinder have been described using a mathematical model. The signals measured by the sensors (pressure, displacement, and airflow) were processed to extract key features, which were then analysed using an SVM classifier to detect anomalies at an early stage. Results The proposed framework showed better fault diagnosis capability in the initial phase of actuator degradation, which further improved the diagnostic accuracy and minimised the potential for unpredictable system failures. Mathematical modelling along with machine learning enabled a reliable condition-monitoring solution which can be used to enable predictive maintenance and enhance the operational reliability and availability of renewable energy systems. Conclusion The proposed framework for predictive maintenance is a practical and sustainable approach that can enhance reliability, efficiency, and lifespan of electro-pneumatic actuators in renewable energy systems. This helps to minimise downtime due to maintenance activities and improves cost-effectiveness.
- 28 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Sustainability-Oriented Predictive Maintenance Framework for Renewable Energy Actuation Systems
Background In renewable energy systems such as wind turbines, solar tracking systems, and energy storage facilities, electromechanical and electro-pneumatic actuation subsystems play a crucial role. These actuator failures lead to lower energy production efficiency, higher maintenance costs, and higher carbon emissions because of unexpected downtime and frequent maintenance. Due to this, there is an increasing demand for intelligent predictive maintenance strategies that can spot incipient faults before they become critical failures. Methods In this study, a framework for predictive maintenance through model-based dynamic analysis and Support Vector Machine (SVM) based incipient fault detection for electro-pneumatic actuation systems in renewable energy applications is presented. The actuator dynamics of a double-acting cylinder have been described using a mathematical model. The signals measured by the sensors (pressure, displacement, and airflow) were processed to extract key features, which were then analysed using an SVM classifier to detect anomalies at an early stage. Results The proposed framework showed better fault diagnosis capability in the initial phase of actuator degradation, which further improved the diagnostic accuracy and minimised the potential for unpredictable system failures. Mathematical modelling along with machine learning enabled a reliable condition-monitoring solution which can be used to enable predictive maintenance and enhance the operational reliability and availability of renewable energy systems. Conclusion The proposed framework for predictive maintenance is a practical and sustainable approach that can enhance reliability, efficiency, and lifespan of electro-pneumatic actuators in renewable energy systems. This helps to minimise downtime due to maintenance activities and improves cost-effectiveness.
- Open Access
- 3 July 2026
- Published by Elsevier in Intelligence-Based Medicine
- Vol. 15
Enhanced detection of malaria parasites in resource constraint environment using data-efficient image transformer
- 3 July 2026
- Published by Elsevier in Intelligence-Based Medicine
- Vol. 15
Open AccessEnhanced detection of malaria parasites in resource constraint environment using data-efficient image transformer
- Open Access
- 26 August 2025
- Published by Innovative Research Publishing in International Journal of Innovative Research and Scientific Studies
- Vol. 8 (6) , 2557-2581
A systematic review on hybrid composite materials for enhancing fuel efficiency and reducing emissions in automobiles
The growing concern over environmental impacts and the need for sustainable practices have prompted research into innovative materials that can improve vehicle performance while minimizing ecological footprints. Hybrid composite materials combine lightweight, high-strength components such as carbon fiber and polymers with metal alloys to reduce vehicle weight, improve fuel efficiency, and lower emissions while maintaining structural integrity and performance in automobiles. Therefore, this systematic review explores the role of hybrid composite materials in enhancing fuel efficiency and reducing emissions in the automotive sector. The review synthesizes existing literature on various hybrid composites, examining their mechanical properties, weight reduction potential, and overall impact on fuel consumption and emissions. Key findings highlight the advantages of integrating multiple materials to optimize performance characteristics and the challenges associated with manufacturing and recycling these composites. The study concludes with recommendations for future research directions, emphasizing the importance of interdisciplinary approaches to develop effective hybrid materials that meet the evolving demands of the automotive industry.
- 26 August 2025
- Published by Innovative Research Publishing in International Journal of Innovative Research and Scientific Studies
- Vol. 8 (6) , 2557-2581
Open AccessA systematic review on hybrid composite materials for enhancing fuel efficiency and reducing emissions in automobiles
The growing concern over environmental impacts and the need for sustainable practices have prompted research into innovative materials that can improve vehicle performance while minimizing ecological footprints. Hybrid composite materials combine lightweight, high-strength components such as carbon fiber and polymers with metal alloys to reduce vehicle weight, improve fuel efficiency, and lower emissions while maintaining structural integrity and performance in automobiles. Therefore, this systematic review explores the role of hybrid composite materials in enhancing fuel efficiency and reducing emissions in the automotive sector. The review synthesizes existing literature on various hybrid composites, examining their mechanical properties, weight reduction potential, and overall impact on fuel consumption and emissions. Key findings highlight the advantages of integrating multiple materials to optimize performance characteristics and the challenges associated with manufacturing and recycling these composites. The study concludes with recommendations for future research directions, emphasizing the importance of interdisciplinary approaches to develop effective hybrid materials that meet the evolving demands of the automotive industry.
- 7 April 2025
- Published by Springer Nature in SN Computer Science
- Vol. 6 (4) , 1-13
Computer Vision and Control Algorithm for Sustainable and Precision Agriculture
- 7 April 2025
- Published by Springer Nature in SN Computer Science
- Vol. 6 (4) , 1-13
Computer Vision and Control Algorithm for Sustainable and Precision Agriculture
- Open Access
- 1 December 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (4) , 4019-4028
Development of a prioritized traffic light control system for emergency vehicles
This research presents a model for an adaptive traffic signal control system aimed at improving urban traffic regulation. It dynamically adjusts signal timing based on vehicle volume at intersections, prioritizing emergency vehicles by allowing them immediate passage. Utilizing Arduino coding, the system controls traffic light intensity according to the traffic flow, enhancing road safety and efficiency. This innovative approach not only facilitates faster clearance for emergency services without human intervention but also reduces congestion and accident rates. This research creates a model for a prioritized traffic signal control system. When the vehicular volume at the intersection varies, the signal time alters autonomously. It identifies the ambulance/emergency vehicles and allows the green light for emergency vehicles like ambulances, and fire engines. This approach may be used to detect traffic accidents and infractions of automobile spiral motions. When erected on the road, the entire system allows for quick traffic clearing for rescue vehicles without requiring a policeman. The system's design eliminates the need for sensors or radio frequency identification (RFID) tags, simplifying traffic management. Simulations validate that emergency vehicle travel time is significantly reduced, proving the system's effectiveness in streamlining urban traffic flows.
- 1 December 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (4) , 4019-4028
Open AccessDevelopment of a prioritized traffic light control system for emergency vehicles
This research presents a model for an adaptive traffic signal control system aimed at improving urban traffic regulation. It dynamically adjusts signal timing based on vehicle volume at intersections, prioritizing emergency vehicles by allowing them immediate passage. Utilizing Arduino coding, the system controls traffic light intensity according to the traffic flow, enhancing road safety and efficiency. This innovative approach not only facilitates faster clearance for emergency services without human intervention but also reduces congestion and accident rates. This research creates a model for a prioritized traffic signal control system. When the vehicular volume at the intersection varies, the signal time alters autonomously. It identifies the ambulance/emergency vehicles and allows the green light for emergency vehicles like ambulances, and fire engines. This approach may be used to detect traffic accidents and infractions of automobile spiral motions. When erected on the road, the entire system allows for quick traffic clearing for rescue vehicles without requiring a policeman. The system's design eliminates the need for sensors or radio frequency identification (RFID) tags, simplifying traffic management. Simulations validate that emergency vehicle travel time is significantly reduced, proving the system's effectiveness in streamlining urban traffic flows.
- Open Access
- 17 September 2024
- Published by Elsevier in Data in Brief
- Vol. 57, 110950
Dataset for a novel AI-powered diagnostic tool for Plasmodium parasite detection authors
- 17 September 2024
- Published by Elsevier in Data in Brief
- Vol. 57, 110950
Open AccessDataset for a novel AI-powered diagnostic tool for Plasmodium parasite detection authors
- Open Access
- 10 September 2024
- Published by Elsevier in Data in Brief
- Vol. 57, 110922
Antimicrobial resistance dataset for pattern recognition in machine learning application
- 10 September 2024
- Published by Elsevier in Data in Brief
- Vol. 57, 110922
Open AccessAntimicrobial resistance dataset for pattern recognition in machine learning application
- Open Access
- 28 May 2024
- Published by MDPI AG in Preprints.org
Impact of a Workplace Screening Programme for Hypertension: A 5-Year Machine Learning-Based Analysis of a University Workforce Medical Records
The burden of hypertension remains unacceptably high globally, particularly in low- and middle-income countries (LMICs). Workplace offer enormous potential as an idea setting for early detection and treatment of hypertension among the working class. Analysis of such a Workplace Screening Programme can reveal information about its potential impact. Machine learning techniques such as k-Means Clustering are untapped tools for such analyses. We set out to deploy this tool for the analysis of our university annual medical screening of workforce for hypertension. An anonymized dataset containing the demography and blood pressure measurements values obtained from staffs of different departments/units in an academic institution was obtained. The total number of samples or data points are 1, 723 in which the input dataset contains six features, including year category (2018, 2019, 2021,2022), Department/Unit (academic and non-academic), gender (male and female), while the target output is the blood pressure status (low, normal and high) respectively. Analysis of the dataset was carried out using machine learning techniques. In this retrospective analysis, it was observed that the mean age for this working class is 42 years old. Similarly, it was discovered that hypertension was prevalent among members of staff above the age of 40 irrespective of their gender or professional category (academic or non-academic). The analysis also revealed that there was a steady decline in the prevalence of hypertension from 2018 to 2022. From the research, it is evident that the adoption of machine learning techniques for periodic analyses of workplace health status screening initiative (especially for hypertension) is effective, feasible, and sustainable to diagnose and control hypertension among the working class.
- 28 May 2024
- Published by MDPI AG in Preprints.org
Open AccessImpact of a Workplace Screening Programme for Hypertension: A 5-Year Machine Learning-Based Analysis of a University Workforce Medical Records
The burden of hypertension remains unacceptably high globally, particularly in low- and middle-income countries (LMICs). Workplace offer enormous potential as an idea setting for early detection and treatment of hypertension among the working class. Analysis of such a Workplace Screening Programme can reveal information about its potential impact. Machine learning techniques such as k-Means Clustering are untapped tools for such analyses. We set out to deploy this tool for the analysis of our university annual medical screening of workforce for hypertension. An anonymized dataset containing the demography and blood pressure measurements values obtained from staffs of different departments/units in an academic institution was obtained. The total number of samples or data points are 1, 723 in which the input dataset contains six features, including year category (2018, 2019, 2021,2022), Department/Unit (academic and non-academic), gender (male and female), while the target output is the blood pressure status (low, normal and high) respectively. Analysis of the dataset was carried out using machine learning techniques. In this retrospective analysis, it was observed that the mean age for this working class is 42 years old. Similarly, it was discovered that hypertension was prevalent among members of staff above the age of 40 irrespective of their gender or professional category (academic or non-academic). The analysis also revealed that there was a steady decline in the prevalence of hypertension from 2018 to 2022. From the research, it is evident that the adoption of machine learning techniques for periodic analyses of workplace health status screening initiative (especially for hypertension) is effective, feasible, and sustainable to diagnose and control hypertension among the working class.