LAUTECH Journal of Engineering and Technology (LAUJET)
Not a member yet
571 research outputs found
Sort by
DEVELOPMENT OF AN INTELLIGENT SURVEILLANCE SYSTEM FOR GESTURE RECOGNITION USING MACHINE LEARNING
Get PDFOver the years, video surveillance systems have been in use in various contexts such as traffic control, crowd control and protecting wildlife. In a dispensation characterised by security concerns and technological advancements, it has become necessary to develop intelligent systems for the purpose of surveillance. Intelligent video monitoring has become a vital tool for boosting security and safety in public areas. These systems combine the use of computer vision, machine learning, and artificial intelligence techniques to analyse video data and alert security personnel to potential threats. They can be configured to recognize particular behaviours, such as loitering, fighting, or unauthorized access to restricted areas, as well as to recognize particular objects, like persons or automobiles. The popularity of these systems has grown recently, owing to their ability to analyse video streams in real time and detect unusual conducts or events. However, traditional manual monitoring methods need a lot of manpower and are easily compromised. In addition, the cost of video surveillance goes up with mass data storage. This research proposes an Intelligent Surveillance System for Gesture Recognition. With sophisticated gesture recognition capabilities, this project focuses on enhancing security measures through cutting-edge technologies. The main objective is to identify potential threats in real-time, concentrating on spotting unusual conduct. The proposed system uses Media pipe to obtain data in real-time. Four different machine learning pipelines were trained for the gesture recognition. They include linear regression, Ridge Classifier, Random Forest Classifier, and Gradient Boosting Classifier, achieving an accuracy of 98.3%, 99.8%, 99%, 98.8%, a precision of 96.8%, 99.8%, 97,7%, 99.2%, a recall of 96.4%, 99.4%, 97.9%, 98%, and an F1-score of 96.5%, 99.6%, 97.8%, 98.6% respectively
Investigating impacts of ammonium phosphate on ash yield from co-combustion of sugarcane bagasse and banana leaves
Get PDFThis study aimed to investigate the influence of mono ammonium phosphate (MAP) on ash yield from the co-combustion of sugarcane bagasse and banana leaves in a muffle furnace. An I-Optimal design of the Combined Methodology, embedded in Design Expert (version 13.0.5), was employed to optimize ash yield, considering various particle sizes, additive concentrations, and temperatures. The mixed samples were ashed in a muffle furnace to a constant weight, and the ash yield was analyzed using statistical tools to assess the model's quality. The MAP additive was most effective at concentrations between 4% and 7%, beyond which ash yields increased significantly. The optimal composition was determined to be 75% sugarcane bagasse, 20% banana leaves, and 5% ammonium phosphate at 950°C, resulting in the lowest ash yield of 6.46%. The presence of MAP in the biomass mixture significantly reduced ash yield. The model's R² and adjusted R² values were 0.9825 and 0.9099, respectively, indicating accurate determination of the model coefficients. This study demonstrates that ammonium phosphate additive has great potential in mitigating ash-related problems in biomass combustion
Co-digestion of Hyptis suaveolens (bushmint) and poultry manure for energy generation: Effects of pretreat-ment methods, Modelling and process parameter optimization study
Get PDFThe potentials of anaerobic co-digestion of poultry dropping with chemically pretreated and untreated Hyptis suaveolens (bushmint weeds) shoots for biogas generation, as well as the process optimization after using a combination of mechanical and thermo-alkaline pretreatment methods, were assessed in this study. Inoculum from cattle rumen content was used to anaerobically digest the shoots. A batch experiment was designed by the Central Composite Design (CCD). Standard procedures were used to assess the physicochemical parameters of the substrates and inoculum, as well as the components of the generated biogas. For the chemically pretreated and untreated tests, the experimental biogas yields were 0.7652 L/kg VS and 1.1396 L/kg VS, respectively. The methane and carbon dioxide content of biogas from both experiments was 69.08%; 17.64% and 62.13%; 21.39% respectively. The Response Surface Methodology (RSM) was employed in data optimization. The predicted biogas yield was 0.7652L/kg VS in the chemically treated experiment, and the model's coefficient of determination (R2) was high (0.9159), indicating strong modeling and prediction accuracy for the chemically treated experiment. There was a 11.19% increase in methane gas yield in the chemically treated experiment over the untreated. The study recommended the worldwide usage of Hyptis suaveolens shoots for biofuels generation and several combinations of pretreatment methods to improve biogas yields
Evaluation of Performance Parameters of Direct Injection Spark Ignition Engines Fuelled with Alcohol-Gasoline Blends
Get PDFAbstract
This study investigates the performance parameters of Direct Injection Spark Ignition (DISI) engines fuelled with alcohol-gasoline blends under varying engine loads and speeds, focusing on methanol, ethanol, butanol and gasoline blends. Alcohols have been recognized for their potential to improve engine performance and reduce harmful emissions due to their higher oxygen content and favourable combustion characteristics. The experimental tests were conducted using a single-cylinder DISI engine under controlled laboratory conditions. Pure gasoline and two alcohol-gasoline blends were tested at different engine loads (0%, 50%, and 100%) and speeds (2500, 3000 and 3500 RPM). D-optimal Response Surface Methodology (RSM) design of Design-Expert version 13.0.1with eleven experimental runs, two factors and seven responses were used for the experimental design and to construct mathematical models for the performance parameters. The adequacy of the models was determined by statistical methods and the best fit model for each response was selected and analysis of variance was applied for a better understanding of model attributes. The results show that alcohol-gasoline blends significantly improve the combustion efficiency of DISI engines. The thermal efficiency of the alcohol-gasoline blend increased slightly by 0.9% compared with that of pure gasoline The addition of alcohols led to an increase in Brake Specific Fuel Consumption (BSFC) by up to 1.89% and engine torque exhibited a marginal improvement. The study concludes that alcohol-gasoline blends can serve as viable solutions to meet future energy efficiency goals in the automotive industry, while also contributing to the reduction of fossil fuel consumption
Optimized convolution neural network-based model for detection and classification of pulmonary diseases.
Get PDFPelican Optimization Algorithm-based Convolutional Neural Network (POA-CNN) method for the automated identification of pulmonary disorders such as COVID-19 and pneumonia is proposed in this research. The method makes use of several processing layers in order to comprehend the representation of stratified data. The three primary phases of the model are feature extraction via POA-based hyperparameter optimization, image classification, and image pre-processing. This approach improves existing systems' performance in detecting pulmonary diseases, highlighting the potential of deep learning in identifying and categorizing human diseases. The study uses a resizing, grayscale, and augmentation method to optimize an existing CNN model. A Convolutional Neural Network (CNN) is then applied to classify Pneumonia and Covid-19 cases. The proposed model achieves an accuracy rate of 97.28 and 97.00%, outperforming existing models. This technique is effective in detecting and classifying other pulmonary diseases, and can be used to automatically detect and classify these diseases. Higher accuracy findings show how successful the model is, making it a useful tool for pulmonary illness identification
Optimization of municipal solid waste management system in landfill using response surface methodology - a case study of Ogbomoso South Local Government
Get PDFIn this study, Solid waste generation and collection assessment within the study area was undertaken with the view of formulating an optimization model that helped identify the important variables influencing the overall cost of solid waste management and also determine their optimum values. Some of the important variables used in formulating the model include; The number of solid waste collection trips (X1), Number of manpower (Persons) (X2), Fuel consumption per day (X3), Weight of solid waste collection per trip (X4).
To determine the model, Statistical design of experiment (DOE) using central composite design method (CCD) was employed. The number of experimental runs based on the CCD method was determined and thirty (30) experimental runs were thereafter generated and optimized.
Result of the optimization model revealed that; the number of solid waste collection trips per month (X1 = 47), Number of manpower (persons) (X2 = 3), Fuel consumption per day (X3 = 18.20liters) and Weight of solid waste collection per trip (X4 = 6.33tons). The optimal solution of selected variables produced an overall cost of N162,300.
The solution was selected by the Design Expert software with a desirability value of 1 that is 100% reliability
A DUAL-TRIGGER SMART HAND WASHING MACHINE
Get PDFHand washing is simple and effective in preventing the transmission of infection and sickness in various contexts, including the home, workplace, childcare facilities, and hospitals. It is important to note that contaminated surface like tap heads and manually operated hand sanitizer pose threat to the users of such facility and as such has constituted a global concern due to the emergence of diseases that can be easily transmitted. Hence, this project presents a low-cost automatic hand washing machine with a temperature sensor and counter, triggered by an Ultrasonic sensor and Laser-Light Dependent Resistor (LDR) trip wire that puts ON or OFF the pump and counting at the same time. The system design was done in two levels, the 3D model and circuit diagram design: firstly, the 3D model design was done with Autodesk Inventor 2017 while the circuit diagram design was done using Fritzing software and simulations performed for both levels. The system was fabricated and evaluated; the result obtained revealed reliable water dispense since the water flow can be activated by either an ultrasonic or laser-LDR sensor, a strong frame at a threshold weight of 166.6 N, and reliable temperature measurement. The uniqueness of this work is that it combines automated temperature measurement, hand-washing, and counting systems in a single device. Temperature measurement and hand-washing help to prevent disease spread while the counting system assists in recording the number of people using or entering a facility to aid the practice of social distancing as means of curtailing the spread of the diseases
OPTIMIZING THE USAGE OF RENEWABLE ENERGY FOR POWERING OFFSHORE OIL FIELDS
Get PDFCarbon dioxide (CO2) emission due to power generation from fossil fuel is a major contributor to the current issues of global warming and climate change. This paper proposes offshore generation of electricity from renewables to supply offshore oil fields. An existing oil field in the UK North Sea was assumed and a hybrid power system consisting of power from wind, wave, and fossil fuel generators was dedicated to it. The feasible/economical reduction in CO2 emission was investigated by using Homer Pro software to model and simulate performance of the micro-grid. Data of the renewable resources are specific to the selected site. From the simulation results, a solution with the lowest net present cost (winning system) was chosen and compared with the base case system to observe how the hybrid system saves cost over the project lifetime. The winning system was refined as much as possible to develop the optimal system which was proposed for implementation. This system demonstrated its economics relative to the base case system as the annual fuel consumption and the corresponding CO2 emission dropped by 38% each. Likewise, the cost of energy fell by 42%. Imposition of carbon tax was recommended to boost the development of renewables
FREQUENCY AND VOLTAGE RESPONSES OF GAS-FIRED DISTRIBUTED GENERATION SYSTEM TO LOAD
Get PDFResearch has shown that addition of loads to a distributed generation (DG) system operating either in stand-alone mode (SAM) or grid-connected mode (GCM) has impacts on its operations. This paper analyzed the voltage and frequency responses of the DG when operated in stand-alone and grid-connected modes under load variations. Mathematical equations showing the characteristics of the DG under varying loads with the two modes were developed. The equations were modeled using MATLAB in Simulink environment. By applying gradual and sudden loads using 2MW DG and an 11kV distribution grid network, the frequency and voltage responses under the two modes were calculated. The results showed that with gradual load addition from 10 to 100 % loading, the output frequency varied from 49.72 to 49.27 Hz (-0.56 to -1.46 %) for SAM while it varied for GCM from 49.90 to 49.44 Hz (-0.20 to -1.12 %). Output voltage varied from 376 to 232.9 V (-6.0 to -41.78 %) for 10 to 100 % load respectively for SAM while it varied from 387.7 to 268.3 V (-3.08 to -32.93 %) for GCM. For sudden load additions, the output frequency variation was between 49.39 to 49.25 Hz (-1.22 to -1.5 %) for 25 to 100 % load for SAM while that of GCM was between 49.51 to 49.43 Hz (-0.98 to -1.14 %); voltage variation was 271.7 to 190 V (-32.08 to -52.5 %) for 25 to 100 % load respectively for SAM while that of GCM was 294.2 to 219.9V(-26.45% to -45.03%). The results revealed that the frequency with gradual and sudden load additions for SAM and GCM varied outside the operational limit of 49.75-50.25 Hz () except in the case of 10 % load under gradual load addition in GCM. However, the frequency and voltage variation are less in GCM than SAM with gradual and sudden load addition
A NON-LINEAR FILTERING APPROACH TO RIDGE AND FURROW SEGMENTATION OF FINGERPRINTS
No full textRidge and furrow segmentation or ridge extraction is an important processing step in automatic fingerprint identification; as its success simplifies the task of tracing the most distinguishing features of the print, the ridge ends and bifurcations. In this work, a new method for ridgeextraction in fingerprints is proposed. The method uses normalisation, local histogram equalisation, median filtering and global thresholding to segment the fingerprint foreground into ridges and furrows. The result obtained shows that the new algorithm is robust to image noise. It is also less computationally demanding when compared with an earlier ridge detection scheme