15288 research outputs found
Sort by
Development of a Sign Language Recognition System Using Machine Learning
Get PDFDeafness and voice impairment have been persistent disabilities throughout history,
hindering individuals from engaging in verbal communication and leading to their isolation
from the predominantly vocally communicating society. Sign language has emerged as the
primary mode of communication for people with these disabilities. However, it presents a
language barrier as it is not commonly understood by those who can hear. To address this
issue, various methods for recognizing sign language have been proposed. This paperaims
to develop a machine learning-based system that can recognize sign language in real-time.
The paper involved the acquisition of a dataset consisting of 44,654 images representing
the static American Sign Language (ASL) alphabet signs. The HandDetector module was
utilized to detect and capture images of the signer's hand forming each sign through a PC
webcam. The dataset was split into three sets: training data (20,772 cases), validation data
(8,903 cases), and test data (14,979 cases). Image pre-processing techniques were
implemented on the images and a convolutional neural network (CNN) model was trained
and compiled. The CNN utilized in the paper comprised of three convolutional layers and a
SoftMax output layer and it was compiled using the Adam optimizer and categorical crossentropy
loss function. The performance of the system was evaluated using the test dataset.
Notably, the system achieved remarkable accuracy rates, having a training accuracy of
99.86%, a validation accuracy of 99.94%, and a test accuracy of 94.68%. The results
obtained from this study demonstrated significant advancements in sign language
recognition, surpassing previous findings in the literature
Impact of material selection on the efficiency of the crushing unit low-medium carbon steel
Get PDFThe quest for sustainability in feed production industries and crushing machines in agricultural sectors that are more efficient have introduced the use of proper material for hammer material. Around the world, the manufacture of feed from animal waste is expanding quickly, and crushing equipment is now essential to enabling reliability. The issue of early failure of the crushing machine’s main components, however, is a challenge for the feed industries and has a direct impact on the machine's maintenance, dependability, and running costs. A significant number of technical components have been created during the past decade for industrial applications employing novel materials and cutting-edge technologies through the development of carburisation. As a result, this review offers a concise summary of the most recent analysis of tribological issues related to crushing hammers made of low and medium-carbon steels. Recent studies on innovative crushing material design, improvement in hammer surface engineering, use of case-hardened hammers with a focus on material selection, crushing machine design optimisation, and failure mode analysis are included in the study. Additionally, it will intricate on the heat treatment technology’s present constraints and its future opportunities
DEVELOPMENT OF AN ANN-BASED DEFECT DETECTION SYSTEM FOR PROCESS QUALITY OPTIMIZATION IN BOTTLING INDUSTRY
Get PDFThe study focused on a novel method for detecting defects in bottle products using a convolutional
neural network (CNN). The convolutional neural network tool was used to capture and analyze defects
during the packaging of beverage products. It involved an Internet of Things (IOT) system that
contains both a client- and server-side system. The client-side system is a raspberry pi, which
captures the bottle sample using its camera and sends it over the Internet of Things (IOT) to the
server for processing. The server processes it using the convolutional neural network and then outputs
the result to the client side for indication. Two convolutional neural network models were developed.
The result showed that the first CNN model detected two states of the bottle, classified as defect and
good state. The second detects up to five defects in a bottle. It was observed that the training process
of CNN model has a prediction accuracy of 90.165% for the first CNN model and 85% for the second
CNN model during live testing. There was observed a positive outcome in the second prediction of
labels the CNN model on most parameters used. Thus, in the development of a defect detection
algorithm, the outcome of this result can form the basis for developing an integrated vision-based
system for tracking defects during the packaging of bottles in the bottling industry. These results
provide potential information and guide for industries on the need to improve their automation in
terms of product defect tracing, thus, improving productivity
Corrigendum to “Experimental investigation of heating values and chemical compositions of selected fuel woods as bio-fuel sources in developing countries”
Get PDFFinancial development and income inequality in Africa
Get PDFAcross the globe, a rise in income inequality has been experienced for the last two decades,
particularly in developing countries. This problem of income inequality poses a challenge to
Africa’s ability to attain the United Nations (UN) Sustainment Development Goals (SDGs) of
reduced inequalities (SDG-10). Against this backdrop, there is a need to harness the potential
of financial development to reduce income inequality in Africa. Therefore, this study empirically
examines how financial development affects income inequality in Africa. Financial development
dimensions, access, depth, efficiency, and stability were considered to achieve the study’s
objective. The study applied the system generalized method of moments (SGMM) to analyse
data and the findings showed that each dimension of financial development had a varying
impact on income inequality. Access, stability and efficiency components of financial development
reduce income inequality, while the depth dimension of financial development
exacerbates income inequality in Africa. Therefore, the study recommends that policymakers
should not neglect other dimensions of finance in facilitating economic development
Natural fibres and biopolymers in FRP composites for strengthening concrete structures: A mixed review
Get PDFFibre-reinforced polymer (FRP) laminates/sheets have been used to retrofit concrete structures. Increased global
awareness of environmental protection needs and recent legislation have propelled researchers to develop more
environmentally friendly FRP materials to be used in place of synthetic FRP materials. Through a bibliometric
and systematic literature review, this article reports on research on the use of bio-based FRP materials comprising
of either natural fibres or biopolymers as external reinforcement for concrete structures. Eighty-seven experi-
mental studies retrieved from Scopus and Google scholar databases were considered for this study. Analysis and
visualization of research output per year and region, re-occurring keywords, co-authorship network and docu-
ment co-citation networks are presented. The effects of various bio-based FRP materials used to strengthen
various concrete members considering different FRP fabrication techniques and FRP configurations are pre-
sented. The study revealed that bio-based FRPs could effectively strengthen concrete beams and columns.
Durability, cost and sustainability of these materials are also discussed. The paper also outlines pathways for
further research and considerations for developing design frameworks
Application of ginger and grapefruit essential oil extracts on the corrosion inhibition of mild steel in dilute 0.5 M H2SO4 electrolyte
Get PDFAdmixture of ginger and grapefruit essential oils (GPP) were studied for their corrosion inhibition properties on mild steel (MS) in 0.5 M H2SO4 solution by potentiodynamic polarization, open circuit potential measurement, electrochemical impedance spectroscopy, weight loss analysis and ATF-FTIR spectroscopy. Results from potentiodynamic polarization shows GPP significantly reduced the corrosion of MS from 8.430 mm/y at 0% GPP concentration to values between 1.979 mm/y and 0.565 mm/y. The corresponding inhibition efficiency values ranged from 76.52% to 93.5% and corrosion current density from 1.88 × 10−4 A/cm2 to 5.36 × 10−5 A/cm2. GPP displayed mixed-type inhibition at all GPP concentrations studied. The OCP plot at 0% GPP initiated at -0.495V compared to -0.443V and -0.451V at 1% and 3.5% GPP. At 9000s, the corresponding OCP values are -0.442V, -0.410V and -0.424V due to electropositive plot shift and passivation of MS surface at 1% and 3.5% GPP, though significant potential transients were present on the OCP plot at 1% GPP. The electrochemical impedance results indicate that the corrosion resistance of MS increased from 4.402 Ω cm2 to 99.318 Ω cm2 upon the addition of 3.5% GPP resulting in inhibition efficiency of 96%. Data from weight loss analysis shows decrease in corrosion rate from 184.48 mm/y to values between 8.94 mm/y and 6.25 mm/y. The corresponding inhibition efficiency values varies from 95.16% at 1% GPP to 96.61% at 3.5% GPP concentration. The ATF-FTIR results confirm the adsorption of GPP molecules on the surface of the carbon steel electrode
Electrochemical data on the corrosion inhibition performance of admixed Citrus paradisi and Zingiber officinale oil extracts in 0.5 M H2 SO 4 solution
Get PDFData output on the protection performance of the combined admixture of Citrus paradisi and Zingiber officinale oil extracts (CPZO) on the corrosion inhibition of mild steel in 0.5 M H2SO4 solution are presented and described. CPZO effectively suppressed the electrochemical reactions on the steel surface with final inhibition values ranging between 95.16 mm/y at 1 % CPZO and 96.61 % at 3.5 % CPZO concentration. CPZO reduced the corrosion rate of the steel from the value of 184.48 % at 0 % CPZO concentration to values between 3.45 mm/y and 8.94 mm/y in the presence of CPZO extract. CPZO performance remained marginally steady with respect to its concentration. However, variation with time was significant till 240 h. Effective inhibition performance was attained at 48 h of exposure. Calculated data from ANOVA analysis shows exposure was the only statistically relevant parameter at 98.63 % compared to 0.75 % for CPZO concentration. Adsorption of CPZO molecules on the steel aligns with the Langmuir and Frumkin isotherm models with correlation coefficient values of 0.9997 and 0.7292. Thermodynamic calculations depict chemisorption adsorption mechanism with ΔG values ranging between −46.13 and −43.23 KJ/mol. Optical images of the inhibited and non-inhibited steel significantly contrast each other
Experimental investigation of heating values and chemical compositions of selected fuel woods as bio-fuel sources in developing countries
Get PDFAgro-waste disposal is a serious environmental problem in developing countries like Nigeria since there are insufficient waste management systems in place. However, it is possible to produce sustainable energy from these biomass wastes, which will lessen environmental damage. The heating value of biomass determines its energy content. The aim of this study was to determine experimentally the higher heating value (HHV) of five selected indigenous fuelwood sawdust and to assess the chemical composition of the pyrolysis yield products using a gas chromatography-mass spectrometry (GC–MS) analyzer. Results of the experimental analysis show that the HHVs of the selected fuel woods: Adansonia digitata (Ad), Terminalia ivorensis (Ti), Khaya ivorensis (Ki), Mansonia altissima (Ma), Okoubaka aubrevillei (Oa) are respectively, 21.02, 20.78, 20.75, 19.95, 19.80 and 20.46 MJ kg−1. According to ultimate analysis-based correlation equation, the HHVs were found to be 18.56, 18.48, 18.42, 18.39 and 18.36 MJ kg−1 for Ad, Ti, Ki, Ma and Oa, respectively. While the proximate analysis-based correlation equation gave HHVs of 18.08, 18.12, 18.25, 18.16 and 18.37 MJ kg−1 for Ad, Ti, Ki, Ma and Oa, respectively. The mean square error (MSE) was used to compare the deviation of the computed results from the experimental data. The statistical analysis indicates comparative agreement between the computed HHVs and the experimental data. The GC–MS analysis shows the presence of phenolic, ketone, fatty acid, ester, and alcohol compounds in the sawdust samples which is evidence that they have chemical and fuel compositions suitable for use as feedstocks in the pharmaceutical and dye industries as well as for the production of biodiesel for internal combustion engines. It can be inferred that the woody biomass residues can be useful sources of biofuels for developing nations' sustainable energy development if adequately processed with suitable technologies