Browsing by Author "Dogo, E. M."

Now showing 1 - 20 of 25

A Comparative Analysis of Gradient Descent-Based Optimization Algorithms on Convolutional Neural Networks
(IEEE, 2018) Dogo, E. M.; Afolabi, O. J.; Nwulu, N. I.; Twala, B.; Aigbavboa, C. O.
In this paper, we perform a comparative evaluation of seven most commonly used first-order stochastic gradient-based optimization techniques in a simple Convolutional Neural Network (ConvNet) architectural setup. The investigated techniques are the Stochastic Gradient Descent (SGD), with vanilla (vSGD), with momentum (SGDm), with momentum and nesterov (SGDm+n)), Root Mean Square Propagation (RMSProp), Adaptive Moment Estimation (Adam), Adaptive Gradient (AdaGrad), Adaptive Delta (AdaDelta), Adaptive moment estimation Extension based on infinity norm (Adamax) and Nesterov-accelerated Adaptive Moment Estimation (Nadam). We trained the model and evaluated the optimization techniques in terms of convergence speed, accuracy and loss function using three randomly selected publicly available image classification datasets. The overall experimental results obtained show Nadam achieved better performance across the three datasets in comparison to the other optimization techniques, while AdaDelta performed the worst.
A decade bibliometric analysis of underwater sensor network research on the Internet of Underwater Things: An African perspective
(Springer, Cham, 2020) Salami, A. F.; Dogo, E. M.; Makaba, T; Adedokun, E. A.; Muazu, M. B.; Sadiq, B. O.; Salawudeen, A. T.
Recent advancements in cloud computing (CC) and the rapid growth of the Internet of Things (IoT) have tremendously revolutionized terrestrial wireless sensor networks (TWSN) communication. These have resultantly paved the way for the practical realization of underwater wireless sensor networks (UWSN) and the emergence of the Internet of Underwater Things (IoUT). The need for better environmental monitoring within the context of smart cities and the recent spate of global natural disasters has further aroused research interest in IoUT which has motivated a number of UWSN innovations, such as the development of tethered remotely operated underwater vehicles (ROUVs), untethered autonomous underwater vehicles (AUVs), unmanned/autonomous surface vehicles (USVs/ASVs) and other smart underwater technologies. While these inventions hold promising prospects for technologically advanced countries, the same assertion cannot be made for most African countries due to challenges inherent in research and development activities into critical IoUT/UWSN projects in the region. This chapter conducts a systematic bibliometric analysis that highlights the knowledge base for core research works in UWSN globally and within the African region. This research discovered 1025 peer-reviewed articles in 5 Scopus-indexed document sources published between 2008 and July 2019. Microsoft Excel and VOSviewer science mapping software tool was used to analyse the retrieved data from Scopus repository. The bibliometric analysis was used to evaluate specific criteria, namely, major subject area, document sources, most cited and productive authors, countries, institutions, funding institutions and most used keywords. The findings of this research indicated that UWSN/IoUT research is still in its infancy in the African region. This chapter concludes by highlighting vital missing links, essential research directions and unique technical recommendations that will be of relevance in helping the successful actualization of IoUT/UWSN research projects in Africa.
A GSM-Based Remote Controlled Poultry Feed Dispensing System Using DTMF
(2016) Ahmed, A.; Olaniyi, O. M.; Dogo, E. M.; James, E.
Poultry rearing for small scale and commercial farmers has made large contribution in food production. With the recent decline in contribution of livestock subsectors to the national economy, there is a need to device a means of making poultry farming convenient, attractive and maximize yield. Automated Feeding is considered very important in poultry production; however, many famers in the tropical regions practice subsistence farming and mostly employ manual poultry feeding. This paper presents the development of a GSM-based remote-control poultry feed dispensing system via Dual Tone Multi-frequency (DTMF) for intermittent control of poultry feed dispensing. Proper knowledge in this area by farmer will help in running effective production and increase yield in meat and egg to maximize profit. The system is multi state fully input dependent system whose change of state can be controlled by a remote control. The developed mechatronic system reduces manpower, saves time and operates efficiently with minimal human involvement in poultry feeding. The system demonstrated practical effort regarding the improvement in performance of existing solid feed dispensing for high yield with minimal human intervention
A Model for an Enterprise Automated RFID-Based Pay and Park System
(ceur-ws.org, 2016) Dogo, E. M.; Ahmed, A.; Adelakun, M. O.
Traffic management is one of the challenging problems in urban cities as vehicle owners look for where to park and queue to pay for rented car parks usually on an hourly basis. Therefore, the choice of a suitable, reliable and flexible architecture for Radio Frequency Identification (RFID) based pay and park system readily comes to mind. It is assumed in this work that the parking lot is already known and secured by the vehicle owner in a closed car park; this paper therefore seeks to address and automate the billing system for enterprise car parks. To achieve this, a reliable and accurate enterprise star topology networked RFID based system, that computes the amount to be paid by a user which is calculated based on the time the user enters and exits the park, and the amount the park owner is charging at a particular point in time is proposed for the automated pay and park system. The system comprises of both software and hardware components integrated together. The developed prototype system is able to grant authorized users access to the park within 30ms after verification and open the barrier in 30ms whenever the emergency button is pressed for safety consideration.
A Sensor-Based Data Acquisition System for Soil Parameters to Determine Suitable Crops
(2023) Abisoye, B. O.; Dogo, E. M.; Umar, B. U.; Mamman, I. Z.
Soil parameters monitoring is significant in sustainable crop and food production. The standard strategy of soil parameters monitoring in developing and underdeveloped nations uses manual labor, resulting in wrong decisions in soil management. Inaccurate measurements due to sensor miscalibration or low sensor quality can lead to incorrect soil management decisions and negatively impact crop yield and environmental sustainability. Due to the mentioned challenges, this work aims to develop a Sensor-based Data Acquisition System for Soil Parameters that will enable users to observe various soil parameters like temperature, humidity, water level and soil pH. The system was developed using the combination of hardware and software components. The hardware component comprises of sensory and processing parts. The study calibrates sensors using known pH, moisture, and temperature values for specific crops to grow in Nigeria. The system will aid farmers in determining suitable crops for their farmland and increasing crop yield. The system collects data through a network of sensors installed in the soil and wirelessly transmits the data to a cloud-based server. The collected data is then analyzed and visualized in through a web-based dashboard, providing farmers with information about the state of their soil. The performance evaluation of the system was carried out using response time and accuracy. The average response time of the system was 4 seconds, and the percentage error for temperature and humidity readings when compared to weather forecast readings were 8.20% and 5.08%, respectively. The results show that the proposed system can provide accurate and reliable measurements of soil parameters and can be easily deployed and operated by small-scale farmers. Using this system can result in improved crop yields, reduced wastage, and better overall efficiency in agricultural operations.
Accessing Imbalance Learning Using Dynamic Selection Approach in Water Quality Anomaly Detection
(MDPI, 2021) Dogo, E. M.; Nwulu, N. I.; Twala, B.; Aigbavboa, C.
Automatic anomaly detection monitoring plays a vital role in water utilities’ distribution systems to reduce the risk posed by unclean water to consumers. One of the major problems with anomaly detection is imbalanced datasets. Dynamic selection techniques combined with ensemble models have proven to be effective for imbalanced datasets classification tasks. In this paper, water quality anomaly detection is formulated as a classification problem in the presences of class imbalance. To tackle this problem, considering the asymmetry dataset distribution between the majority and minority classes, the performance of sixteen previously proposed single and static ensemble classification methods embedded with resampling strategies are first optimised and compared. After that, six dynamic selection techniques, namely, Modified Class Rank (Rank), Local Class Accuracy (LCA), Overall-Local Accuracy (OLA), K-Nearest Oracles Eliminate (KNORA-E), K-Nearest Oracles Union (KNORA-U) and Meta-Learning for Dynamic Ensemble Selection (META-DES) in combination with homogeneous and heterogeneous ensemble models and three SMOTE-based resampling algorithms (SMOTE, SMOTE+ENN and SMOTE+Tomek Links), and one missing data method (missForest) are proposed and evaluated. A binary real-world drinking-water quality anomaly detection dataset is utilised to evaluate the models. The experimental results obtained reveal all the models benefitting from the combined optimisation of both the classifiers and resampling methods. Considering the three performance measures (balanced accuracy, F-score and G-mean), the result also shows that the dynamic classifier selection (DCS) techniques, in particular, the missForest+SMOTE+RANK and missForest+SMOTE+OLA models based on homogeneous ensemble-bagging with decision tree as the base classifier, exhibited better performances in terms of balanced accuracy and G-mean, while the Bg+mF+SMENN+LCA model based on homogeneous ensemble-bagging with random forest has a better overall F1-measure in comparison to the other models.
Artificial intelligence model for prediction of cardiovascular disease: An empirical study
(AccScience Publishing, 2024) Umar, B. U.; Ajao, L. A.; Dogo, E. M.; Ajao, F. J.; Atama, M.
Cardiovascular disease (CVD) is a disease related to the heart and blood vessels. Prediction of CVD is essential for early detection and diagnosis, which is however compounded by the complex interplay between medical history, physical examination outcomes, and imaging results. While the existing automated systems are fraught with the usage of irrelevant and redundant attributes, artificial intelligence (AI) helps in the identification of potential CVD populations by prediction models. This work aims at developing an AI model for predicting CVD using different classifications of machine learning techniques. The CVD dataset was obtained from the UCI repository containing about 76 cardiac attributes for training in various machine learning models, which include a hybrid of artificial neural network genetic algorithm (ANN-GA), artificial neural network, support vector machine (SVM), K-means, K-nearest neighbor (KNN), and decision tree (DT). The performance of the models was measured in terms of accuracy, means square error, sensitivity, specificity, and precision. The results showed that the hybrid model of ANN-GA performs better with an accuracy of 86.4%, compared to the SVM, K-means, KNN, and DT measured at 84.0%, 59.6%, 79.0%, and 77.8%, respectively. It was observed that the system performs better as the number of datasets increases in the database, with a fewer selection of attributes using genetic algorithm for selection. Thus, the ANN-GA model is recommended for CVD prediction and diagnosis.
Automatic Photovoltaic Solar Panel Dust Cleaning System
(IGI Global, 2021) Shibane, N.; Nwulu, N.; Dogo, E. M.
Renewable energy sources are currently regarded as viable options for stabilizing the energy crisis globally as well as addressing global warming challenges. Solar energy is the most promising and sustainable energy source as compared to other renewable energy sources such as coal, nuclear, wind, gas, and hydro energy. The increasing demand for solar panels should be reason enough to investigate ways in which we can increase their efficiency as much as possible. Dust, dirt, and bird dropping are major factors that can affect the performance of solar panel systems. This work presents the development of a solar panel cleaning system that automatically detects dust particles and cleans the solar panel to ensure the continues efficiency of the solar system is at an optimal level. The system comprises of five subsystems: dust sensing, water pumping, microcontroller, cleaning mechanism, and the power system. Tests carried out on the system shows its quick response to signals and effectiveness in cleaning the solar panel whenever dust particles are detected.
Blockchain for securing electronic voting systems: a survey of architectures, trends, solutions, and challenges
(Springer, 2025) Ohize, H. O.; Onumanyi, A. J.; Umar, B. U.; Ajao, L. A.; Isah, R. O.; Dogo, E. M.; Nuhu, B. K.; Olaniyi, O. M.; Ambafi, J. G.; Sheidu, V. B.; Ibrahim, M. M.
Electronic voting (e-voting) systems are gaining increasing attention as a means to modernize electoral processes, enhance transparency, and boost voters’ participation. In recent years, significant developments have occurred in the study of e-voting and blockchain technology systems, hence reshaping many electoral systems globally. For example, real-world implementations of blockchain-based e-voting have been explored in various countries, such as Estonia and Switzerland, which demonstrates the potential of blockchain to enhance the security and transparency of elections. Thus, in this paper, we present a survey of the latest trends in the development of e-voting systems, focusing on the integration of blockchain technology as a promising solution to address various concerns in e-voting, including security, transparency, auditability, and voting integrity. This survey is important because existing survey articles do not cover the latest advancements in blockchain technology for e-voting, particularly as it relates to architecture, global trends, and current concerns in the developmental process. Thus, we address this gap by providing an encompassing overview of architectures, developments, concerns, and solutions in e-voting systems based on the use of blockchain technology. Specifically, a concise summary of the information necessary for implementing blockchain-based e-voting solutions is provided. Furthermore, we discuss recent advances in blockchain systems, which aim to enhance scalability and performance in large-scale voting scenarios. We also highlight the fact that the implementation of blockchain-based e-voting systems faces challenges, including cybersecurity risks, resource intensity, and the need for robust infrastructure, which must be addressed to ensure the scalability and reliability of these systems. This survey also points to the ongoing development in the field, highlighting future research directions such as improving the efficiency of blockchain algorithms and integrating advanced cryptographic techniques to further enhance security and trust in e-voting systems. Hence, by analyzing the current state of e-voting systems and blockchain technology, insights have been provided into the opportunities and challenges in the field with opportunities for future research and development efforts aimed at creating more secure, transparent, and inclusive electoral processes.
Bluetooth Assisted Misplaced Object Finder Using DFRobot Arduino Integrated with Android Application
(2024) Dogo, E. M.; Emeni, B.; Nuhu, B. K.; Ajao, L. A.
Finding lost or misplaced items can be time-consuming and frustrating. Yet, this is common and occurs to many individuals daily and globally. This paper has developed a system that allows users to locate their misplaced or lost items by leveraging the capabilities of Bluetooth technology and a microcontroller-based control system. The DFRobot Bettle BLE Arduino microcontroller is the main component for communication and control. By interfacing the microcontroller with an LED and a buzzer, the system provides visual and auditory signals to assist in locating the target device or item. The search pro-cess is initiated through an Android application, through establishing a Blue-tooth connection between the microcontroller and the target device, permitting the exchange of signals for tracking purposes. When the device is within range, the LED indicator illuminates, and the buzzer produces audible alerts, guiding the user to the device’s location. The application also provides an estimated distance of the object using Bluetooth signal strength. Tests carried out on the system proved its effectiveness in terms of quick response to signals and reliability in both indoor and outdoor environments.
Cross-Layer Integration Approach for Improving QoS for IPv6 Based VOIP
(iJET, 2014) Dogo, E. M.; Ahmed, A.; Olaniyi, O. M.
Voice over IP (VOIP) is today one of the most innovative IP based Communication Technologies in the Telecommunications industry. This has made it to enjoy a high degree of success in its application in small, medium and large scale enterprises, primarily to save cost as well as leveraging on its enhance functionalities such as mobility and scalability. Despite all its successes, VOIP still faces challenges with Quality of Service (QoS) degradation. This paper proposes a cross-layer model to effectively manage interactions in the data, network and transport layers guided by tradeoff between three performance metrics that affect QoS of VOIP for an improved QoS for Voice over IPv6 (VOIPv6). The parameters taken into consideration in this proposed model are: packet loss, delay and throughput observe by the end user.
Dataset for a wireless sensor network based drinking-water quality monitoring and notification system
(Elsevier, 2019) Sithole, M. P. P.; Nwulu, N. I.; Dogo, E. M.
This paper presents the collected experimental data for water quality monitoring which was conducted in ten experiments by using five different common sources of water contaminants namely soil, salt, washing powder, chlorine and vinegar and their combination. The data were collected indoors at room temperature during the day for several days using sensors that measure pH, turbidity, flow rate, and conductivity in water. The water consumption risk (CR) was calculated as deviation based on the water quality parameters standards proposed by the World Health Organisation (WHO) and the South African Department of Water Affairs (DWA), with respect to the sensor measurement readings obtained. While the error measurements were calculated based on the expected parameter measurement per conducted experiment and repeated for 26 measurements. Pure tap water was the benchmark of water safe for human consumption. The first five experiments were performed by introducing each contaminant into the water and thereafter, two contaminants in the sixth experiment and their additions until all different contaminants were experimented at once in the last experiment.
Design of a Simple and Low-Cost Microcontroller Based Medicare Device for Heart Beat Monitoring
(IEEE, 2013) Dogo, E. M.; Sado, F.; Adah, S. M.
Heart beat monitoring is vital to ensuring healthiness of the human cardiovascular system, but availability of a simple and low-cost heart beat monitoring device that does not require expert medical personnel to handle still remains a challenge especially in rural and semi urban areas of developing countries like Nigeria. This paper describes the design and implementation of a simple, reliable, accurate and cost effective microcontroller based heart beat monitoring device with Liquid Crystal Display (LCD) and voice outputs. The heart rate of the subject is measured from the fingertip using optical sensors and the rate is then displayed on a text based LCD and voice outputs in English language and two Nigerian indigenous languages (Hausa and Yoruba).
Development and Implementation of Microcontroller-based Improved Digital Timer and Alarm System
(2016) Ajao, L. A.; Adegboye, M. A,; Dogo, E. M.; Aliyu, S. O.; Maliki, D.
Time plays an important role in our daily activities, more particularly in sectional events or conference arena where there is need for accurate time management. This paper focuses on the development and implementation of an improved digital timer with audio-visual unit using (PIC16F887) microcontroller chip and other electronics component such as LCD, 7-segment display, LED and buzzer as an I/O device. Thus, the need for this device in our daily activities is to monitor the time scheduled for events, updating and alert the audience using an audio-visual approach. The proposed system allows apt time management and avoids time wastage during seminar presentations and the likes. It particularly helps presenters to be time conscious, thus, making them to naturally adjust such that the allotted time is enough to cover up their presentation. The digital timer and alarm system presented herewith is also of advantage to the physically challenged like the deaf and blind in monitoring their sectional activities and to be fully involved about the event situation. The system was designed in different modules, and all were interfaced together with firmware chip to simplify the mechanism’s fault diagnoses and fault corrections.
Development of feedback mechanism for microcontroller based SMS electronic strolling message display board
(2014) Dogo, E. M.; Akogbe, A. M.; Folorunso, T. A.; Maliki, D.; Akindele, A. A.
Empirical Comparison of Approaches for Mitigating Effects of Class Imbalances in Water Quality Anomaly Detection
(IEEE, 2020) Dogo, E. M.; Nwulu, N. I.; Twala, B.; Aigbavboa, C. O.
Imbalanced class distribution and missing data are two common problems and occurrences in water quality anomaly detection domain. Learning algorithms in an imbalanced dataset can yield an overrated classification accuracy driven by a bias towards the majority class at the expense of the minority class. On the other hand, missing values in data can induce complexity in the learning classifiers during data analysis. These two problems pose substantial challenges to the performance of learning algorithms in real-life water quality anomaly detection problems. Hence, the need for them to be carefully considered and addressed to achieve better performance. In this paper, the performance of a range of several combinations of techniques to deal with imbalanced classes in the context of binary-imbalanced water quality anomaly detection problem and the presence of missing values is extensively compare. The methods considered include seven missing data and eight resampling methods, on ten different learning state-of-the-art classifiers taking into account diversity in their learning philosophies. The different classifiers are evaluated using stratified 5-fold cross-validation, based on three performance evaluation metrics namely accuracy, ROC-AUC and F1-measure. Further experiments are carried out on nineteen variants of homogeneous and heterogeneous ensemble techniques embedded with resampling and missing value strategies during their training phase as well as an optimized deep neural network model. The experimental results show an improvement in the performance of the learning classifiers, especially when dealing with the class imbalance problem (on the one hand) and the incomplete data problem (on the other hand). Furthermore, the neural network model exhibit superior performance when dealing with both problems.
Food Safety 4.0: The Future of Food Safety Leveraging Industry 4.0 Technologies
(Springer, 2024) Dogo, E. M.; Bokaba, T.; Makun, H. A.; Aliyu, A.; Kparbong, P. B.
The convergence of several factors, including population growth, scarce natural resources, climate change, globalization, sustainability, and advances in modern and emerging technologies, in addition to growing demand by consumers, retailers, regulators, and stakeholders to raise the level of food protection is increasingly driving the global food safety agenda. Industry 4.0 technologies are fast infusing into all fields of human endeavor including the food safety ecosystem, serving as catalysts for innovation and sustainability. However, there is no empirical evidence regarding the extent of their application and the level of acceptance within the food safety domain. This study explored the relationship between Industry 4.0 technologies and food safety by evaluating the applications of the Fourth Industrial Revolution technologies in addressing food safety challenges. The objectives are achieved using qualitative methodology and bibliometric analysis of content. Our analysis indicates that artificial intelligence, the Internet of Things, machine learning, and big data are prominent topics related to the 4IR and food safety, while blockchain and smart manufacturing are emerging topics.
Impact of Gaussian Noise on the Optimization of Medical Image Registration
(2024) Sokomba, A. Z.; Dogo, E. M.; Maliki, D.; Abdullahi, I. M.
Gaussian noise often poses a significant challenge to medical image registration, impacting the accuracy and reliability of alignment across varying imaging modalities. The research investigates the effect of Gaussian noise on medical image registration by comparing four optimization techniques: a direct approach, an optimization using FMINCON, a multiscale approach, and a combined optimization strategy that integrates FMINCON and the multiscale approach. The comparative analysis assesses each method's robustness against Gaussian noise, evaluating registration accuracy through three key similarity metrics: Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM). The results reveal that while each approach demonstrates a degree of resilience to noise, the combined optimization method significantly outperforms the others, achieving the lowest MSE, highest PSNR, and superior SSIM. These findings suggest that the combined approach effectively enhances the optimization process by leveraging the strengths of both FMINCON and multiscale frameworks, thus providing a more accurate and noise-resistant solution for medical image registration. The analysis highlights the necessity of image filtering techniques to mitigate noise interference and improve the image registration process in clinical applications.
Investigating the Thresholding Effect and Fingerprint Transformation Using Cross-Correlation Similarity Matching
(2024) Garuba, O. R.; Abdullahi, I. M.; Dogo, E. M.; Maliki, D.
This research presents a cross-correlation similarity matching method to study the fingerprint transformation and thresholding impact. This work directly compares the impact of various transformations (rotation, translation, elastic deformation, and scaling) on the fingerprint matching performance at different threshold values, in contrast to the standard minutiae-based systems. In order to compare the template positions of the two fingerprints using plots, the cross-correlation similarity matching of fingerprints first selects suitable templates in the primary fingerprint and then uses template matching to assess the impact of each transformation on matching accuracy, FRR, and FAR in the secondary print. The findings highlight the potential of thresholding in developing reliable and practical fingerprint recognition systems.
IoT Based Security Management Framework for Heterogeneous Network Environment
(2020) Ajibo, C. A.; Chinaeke-Ogbuka, I. M.; Dogo, E. M.; Ogbuka C. U.
In an effort to curb the potential losses associated with the event of security bridge, admitting the uneven bandwidth support that characterizes most developing smart cities, we propose a neural inspired Multimodal Security Management System (MSMS) that is bandwidth-tolerant. The proposed system leverages on a Next-Generation Network (NGN) architecture in catering for the challenges associated with the provisioning of ubiquitous broadband access for IoT support in a heterogeneous morphological network environment. In order to evaluate the MSMS, we simulated the proposed cloud-based system on a Next Generation Network (NGN) architecture which utilizes Internet Protocol/Multi-Protocol Label Switching (IP/MPLS) as transport technique in a Long Term Evolution (LTE) backbone infrastructure. We then compare its performance over a competitive alternative transport technique: "Internet Protocol Asynchronous Transfer Mode (IP/ATM)". Thus, we further evaluated the MEMS on the latter architecture. While, our proposed system is able to capture both textual, aural, and visual information of individuals in security vulnerable environments via installed smart microphones and cameras, it is also able to integrate this information's in predicting security threats. When compared with the popular Security Management System (SMS) "ShotSpotter", results show that our proposed system outperforms the ShotSpotter system by 0.87 and 0.45 in terms of efficiency and response time respectively. Finally, simulation of our proposed system on an IP/MPLS transport schemes shows that the former outperforms the latter with respect to overall network bandwidth utilization and average traffic loss in the ratio of 0.098 and 0.087 respectively.