13474 research outputs found
Sort by
Lightweight one-stage object detection models for machine vision based recognition system / Mohamad Haniff Junos
No full textObject detection is an important research area of computer vision and has been applied in numerous real-world applications. Object detection is a complicated task requiring locating and identifying targets with significant variations in shape, size, viewpoint, high obscurity, and complex backgrounds. Many approaches have been developed to address these challenges with a remarkable increase in data availability and computational resources. While current state-of-the-art one-stage object detection methods achieve high detection accuracy, they are unsuitable for embedded devices due to their complex structure and extensive network parameters. Therefore, this study proposes two lightweight object detection models based on the You Only Look Once (YOLO) tiny model to improve detection accuracy, reduce the model size, and achieve real-time performance. For the proposed model I: YOLO-A model, a hybrid backbone structure based on densely connected neural network (DesneNet) and mobile inverted bottleneck module (MBConv) were adopted into the YOLOv3 tiny model for better feature reuse and to reduce the network’s parameters. Then, the swish activation function replaced the Leaky ReLU function to enhance the expression of input data and weight to be learnt. Besides, four detection layers were used to improve object detection of various sizes. Then, a spatial pyramid pooling (SPP) structure was adopted to increase the receptive field in the network, and a complete intersection over union (CIoU) function was used for bounding box regression. The proposed model II: YOLO-S adopted several improvements, including the integration of MBConv into the YOLOv4 tiny model, SPP in the neck section and three detection layers. The developed models were evaluated on the Palm and VisDrone dataset, and the results were compared to several state-of-the-art models in terms of detection performance. The experimental results indicate that the developed models achieved significant improvement over the original YOLO tiny model. On the Palm dataset, the experimental results demonstrated that the proposed YOLO-A model outperformed the state-of-the-art models with an mAP of 97.29%, while the YOLO-S model achieved better accuracy than the other lightweight models with an mAP of 96.41%. Conversely, on the VisDrone dataset, the proposed YOLO-A model obtained slightly lower mAP (36.15%) than the YOLOv4 model but higher than the rest. This is due to the VisDrone dataset, consisting of many challenging images. Besides, the YOLO-S model achieved better mAP than the other lightweight models with 24.73%. The detection speed and computational performance were evaluated on the VisDrone dataset. Results show that both models significantly improve the computational performance where the YOLO-A model generated a model size of 65.6 MB with BFLOPs of 28.416, while the YOLO-S model produced the smallest size with 13.3 MB and BFLOPs of 6.280. Furthermore, both models achieved better detection speed with 3.4 FPS and 17.3 FPS when tested on the Jetson Nano. The proposed models offer the best trade-offs in detection accuracy, model size, and detection time. The comprehensive results exhibit the superiority of the proposed models over several existing state-of-the-art detection models and have the potential to be effectively deployed on the limited capacity embedded devices
Prioritisation assessment and robust predictive model for a comprehensive medical equipment maintenance using machine learning techniques / Aizat Hilmi Zamzam
No full textMedical equipment reliability is critical to the quality of healthcare services. Nevertheless, maintaining the reliability of medical equipment in terms of availability, durability, safety, and economical is a challenging mission. A comprehensive and cost-effective medical equipment maintenance management covering preventive maintenance, corrective maintenance, and replacement plan is needed to achieve the three goals. The study aims to develop a comprehensive strategic maintenance management for sustaining the medical equipment reliability in a cost-effective way. Data such as maintenance history and inventory information on 13,350 units of medical equipment located in health clinics in fourteen states throughout Malaysia were used as samples. The datasets are established according to nineteen features and criteria for this study. The development of predictive models for objectives 1 and 2 of this study involves the application of seven supervised machine learning algorithms. The effectiveness of these models is assessed through eleven performance evaluation parameters. Classifiers that produce the best models are selected for the optimisation process. The optimal models are produced by adjusting the selected classifiers' hyperparameters to reduce the misclassification rate during the prediction process. The achievement of objective 1 demonstrates that the SVM, DT, and NN classifiers have developed optimised predictive models for first failure, failure to year ratio, and failure rectification action, respectively. Meanwhile, the development of predictive models to achieve research objective 2 involves two techniques for assessment of maintenance priorities, namely k-means and classification. The results of these prioritisation assessment techniques are then applied in the development of predictive models. A comparison of the effectiveness demonstrates that he combination of k-means and the NN classifier has developed optimised predictive models for all maintenance management activities at an accuracy rate of over 99.5%. The development of a comprehensive strategic maintenance management includes the elements of maintenance prioritisation and failure analysis to achieve objective 3. It involves the rationalisation of priorities for preventive maintenance, corrective maintenance, and replacement plan predictive models. Moreover, this rationalisation is combined with a first failure analysis prediction, which involves the adjustment of the frequency of planned preventive maintenance and maintenance costs. Integration between rationalisation and a combination of elements shows a reduction in preventive and corrective maintenance costs through the implementation of cost analysis. The results of the analysis found that a 61.4% cost-saving was obtained from the current maintenance costs. This cost-saving can cover 10% of the total estimated cost of procurement of obsolete equipment. This percentage is equivalent to 1,982, which is 40% of the total obsolete equipment proposed for replacement. The establishment of a comprehensive maintenance management through a combination of failure analysis and maintenance prioritisation predictive models can be a mechanism for the implementation of predictive maintenance. Furthermore, it also serves as a tool for clinical engineers in implementing more effective and efficient medical equipment maintenance management
Development of sugarcane bagasse-based adsorbents for dye and antibiotic removal from contaminated water / Farahin Mohd Jais
No full textWater contamination from dyes and antibiotics has been a growing cause of environmental concern. Versatile adsorbents that can remove various contaminants simultaneously rather than just one type of contaminant would be beneficial in developing practical strategies to remove contaminants. Sugarcane bagasse (SB) has the potential to be a versatile, low-cost adsorbent due to its high cellulose, hemicellulose, and lignin content. Hydrothermal carbonization (HTC) is a moderate-temperature alternative to pyrolysis, producing green and cost-effective SB-based adsorbent (hydrochar). Despite this, hydrochar recognize for having poor porosity, which reduces its adsorption efficiency. The activation/modification of SB hydrochar may mitigate its drawbacks. However, difficulties in recovering powdered SB-based adsorbent from the solution during the post-adsorption process may prevent wide application. Hence, the ultimate encapsulated SB-based adsorbent was developed through four levels of sequence development ((1) SB hydrochar (HC), (2) activated HC (AHC), (3) modified AHC by NiFe bimetal organic framework (NiFeMOF) (NiFeMOF@AHCop), and the (4) double crosslinking of encapsulated NiFeMOF@AHCop in alginate bead aerogel (NiFeAHC/SA bead aerogel)) were examined. Response Surface Methodology-Central Composite Design (RSM-CCD) was employed to optimize the preparation parameters of HCop, AHCop, NiFeopAHC, and NiFeCop bead aerogel in achieving maximum dye and antibiotic removal, with Crystal Violet (CV) dye and Tetracycline (TC) as model contaminants. Characterization analyses including Field Emission Scanning Electron Microscopy (FESEM), Fourier-transform infrared spectroscopy (FTIR), Brunaue
Critical success factors in the development of Malaysian Islamic marriage management system / Salahuddin Hamzah
No full textThe aim of this research is to develop a framework depicting the critical success factors in developing an IT system in JAKIM and JAWI. Using quantitative methodology by distributing questionnaire to selected 63 officers and support staff of JAKIM and JAWI, the results were analysed using Relative Importance Indicator. The finding from this research shows that the dominant critical success factor is “human management” factor. Although system development project management is a highly technical project, the “human factor” in project management exerts the most important factor. The importance of leadership/top management or the “human aspect” in project management has been repeated by many researchers. The finding of this research have proven that in a hierarchical governmental organisation, the human aspect of leadership/top management, combined with user’s competencies and the availability of financial resources are the critical factors needed for success. This research has contributed in the understanding of officers and support staff of JAKIM and JAWI in acknowledging the critical factors that contributed to the success of the development of the system. Further improvement with regard to the aspect of top management, leadership and human management can be incorporated in the area of management training and career development of current and future government officers and not limited to those who are in project management related fields, but also other fields of management as well
Development of europium-doped calcium silicate coating for bone tissue engineering applications / Kee Chia Ching
No full textCalcium silicate (CaSiO3 or CS) ceramics are promising bioactive ceramics for coatings on metallic implants (titanium and its alloys) for bone regeneration to overcome their inert properties. Yet, their rapid dissolution rates lead to poor chemical and mechanical stability of the implants. Thus, chemically modified CS ceramics are developed to overcome their drawbacks without impairing their bioactivity. Europium (Eu) ions have been identified as promising doping candidates into bioactive ceramics to achieve different biological and functional properties. In this research, Eu–CS coatings with improved mechanical properties, lower dissolution rate and enhanced bioactivity were successfully fabricated for potential application in bone tissue engineering.
The precursor of CS, calcium silicate hydrate (CSH), doped with various amount of Eu, was synthesized via hydrothermal and coprecipitation methods. Subsequently, both synthesis methods were compared. The hydrothermal method produced xonotlite while coprecipitation gave 11 Å tobermorite. Regardless of the synthesis method, incorporation of Eu inhibited the crystallite growth and particle size of the as-synthesized powders, while increasing their thermal stability. In both phases, Eu3+ was found to occupy the cation sites in preference to Ca2+. Comparing both synthesis methods, the hydrothermally synthesized powders were of higher crystallinity and thermal stability than the powders prepared by coprecipitation. Moreover, Eu–CSH powder by hydrothermal exhibited stronger photoluminescence intensity than the one by coprecipitation, primarily attributed to its higher degree of crystallinity. Thus, they showed higher potential for nanomedicine application where a combination of biocompatibility and light emission is desired. Powders of Eu–CSH by hydrothermal method were electrophoretically deposited (EPD) on titanium substrates, and calcined, in order to obtain Eu–CS coatings. Their chemical, structural, mechanical and biological properties were evaluated. Doping Eu into CS altered the particle shape and morphology of the coatings, where a low amount of Eu gave a relatively denser coating, while a higher amount gave a more porous one. An Eu doping of 2.5 mol% gave the optimal coating adhesion strength. All Eu–CS coatings provided sufficient apatite forming ability, yet lower degradation rate, as compared to CS coating. Eu2.5CS induced formation of calcium-deficient apatite with a Ca/P ratio of 1.43, close to biological apatite with a ratio of ~1.7. The effect of Eu doped CS on human fetal osteoblast (hFOB) cells were investigated in vitro as well. It was observed that hFOB cells could attach and proliferate on all Eu–CS coatings, suggesting the low or absent cytotoxicity of the coatings. A low amount of Eu doping (2.5 mol%) not only showed comparable cell proliferation with CS, but also enhanced the osteogenic activity of the CS coating
Mechanical and thermo-electrical properties evaluation of helical multiwall carbon nanotube enhanced carbon composite / Ali Naem S Alamry
No full textThe introduction of carbon nanotubes (CNTs) inside a polymer matrix is preferred as a reinforcing medium to enhance the mechanical, physical, and electrical properties of composite materials. It was therefore the purpose of this study to experimentally investigate the role of helical multi-walled carbon nanotubes (HMWCNTs) on the fracture behavior of polymer composites and understand the underlying failure mechanisms under monotonic and impact loading conditions. Nanocomposite laminates were manufactured and subjected to various mechanical, thermal, and electrical tests to evaluate their properties. This was followed by a detailed examination of the fracture surface with the help of a field emission scanning electron microscope (FESEM) and ultrasonic C-scanning to investigate the underlying fracture and damage mechanisms. It was found that there was an increase in the mechanical strength of the nanocomposite laminates with the addition of HMWCNTs in comparison with the control sample. From the morphology analysis of fractured surfaces by FESEM, it was also found that upon increasing the number of HMWCNTs in the composite laminate, a higher bridging effect was achieved, which enhanced the strength of the composite laminates. Evidence that fewer HMWCNTs aggregated over the fracture surface was also found in this work
Thermal treatment approach in immobilizing silver nanoparticles into polyvinylidene fluoride membranes for biofouling mitigation during water treatment / Nur Diyana Aqilah Kamarudin
No full textImpregnation of the noble metal, silver nanoparticles (AgNPs) as a disinfectant into the polymeric membrane is one of the empirical modification methods for producing safe drinking water. However, the dispersion and immobilization of AgNPs during membrane fabrication and application are least discussed despite numerous leaching phenomena reported. Moreover, the synthesis method and parameter are crucial factors in determining nanoparticles formation and nucleation process. Additionally, analysis of membrane fouling despite the mitigation approach is vital for an explicit understanding of fouling occurrence. This study presents the effects of precursor concentration and synthesis temperature on AgNPs formation in two distinctive methods. Silver nanoparticles type 1 (AgNPs-1) and type 2 (AgNPs-2) were prepared and incorporated into polyvinylidene fluoride (PVDF) ultrafiltration membranes with the presence of thermal treatment. The effect of thermal treatment was studied at temperatures 25, 40, and 55°C which significantly influenced the physicochemical characteristics of the PVDF membranes. During filtration using lake water as feed, flux decay rate (FDR), and fouling recovery ratio (FRR) analysis of clean and fouled membranes have implied membrane fouling phenomena. Alternatively, the AgNPs incorporated in PVDF membranes were utilized for surface-enhanced Raman spectroscopy (SERS) as an unconventional approach of built-in surface enhance substrate whilst thermogravimetric analysis (TGA) was employed to quantify foulant as an approach to evaluate the performance of AgNPs. The results have shown a significant variation of AgNPs-1 and AgNPs-2 synthesized with different synthesis parameters for both methods. With respect to particles size determined using a particle size analyzer, the best synthesis condition for AgNPs-1 was at 25°C using 0.255 g AgNO3 while on the other hand the best synthesis condition for AgNPs-2 was at 75°C using 0.03 mol/L AgNO3. Membranes incorporated with selected AgNPs-1 and AgNPs-2, fabricated at the highest temperature portrayed improved mechanical properties and pure water flux. PVDF/PVP/AgNPs-2 membrane released the smallest amount of Ag (3.81%) while PVDF/PVP/AgNPs-1 cast at 40°C released the largest amount of Ag (42.34%) during the phase inversion at fabrication. Notably, no Ag leaching was found in all membranes during the dead-end filtration at 0.5 bar. Membranes with AgNPs-1 and AgNPs-2 showed remarkable bactericidal performance towards E. coli with a percentage reduction of 90.38% and 80.77%, respectively. Thus, membranes with AgNPs fabricated at 55°C offered better mechanical properties of the enhanced permeability and excellent stability of AgNPs immobilization which signified the thermal treatment approach. Remarkably, small significant peaks representing biofoulant presence were obtained from normal Raman while total foulant weight indicated reductions of 82.69% and 75.86% with the presence of AgNPs-1 and AgNPs-2, respectively. Hence, the unconventional approach in fouling analysis for AgNPs in PVDF membrane has potentially been earmarked for surface enhancement in normal Raman analysis while foulant quantification provided a profound understanding of fouling occurrences on the membrane which contributes to the development of the fouling mitigation approach
Effectiveness of dengue awareness calendar among Orang Asli population in Selangor / Arulvani Rajandra
No full textBackground: Dengue cases are increasing at an alarming rate and prevalent among the marginalized minority community of Malaysia, Orang Asli due to their impoverished condition and lack of health awareness. Tailoring interventions using health educational materials is expected to curb dengue episodes by improving their knowledge, belief and practices (KBP) on dengue. Aims: This study aimed to evaluate the effectiveness of dengue awareness calendar on knowledge, beliefs, and practices on dengue. This study also investigated the socio-demographic, environmental factors, increment in knowledge score, changes in health belief associated with increment in dengue prevention practices score. Method: A pre and post non control group study was conducted in selected nine Orang Asli villages in Selangor from July 2017 to July 2018. During the interphase of the intervention, a printed dengue awareness calendar on dengue distributed to the participants. The KBP scores on dengue were assessed before and after the interventions in all the nine villages to evaluate the dengue awareness calendar. Results: A total of 609 complete responses were obtained. The dengue awareness calendar significantly improved the knowledge score, perceived severity, cues to action, self-efficacy and the prevention practices score(p>0.05). The increment in prevention practices is significantly associated with tribes, education level, type of occupation, difference in perceived severity, difference perceived susceptibility and increase in total knowledge score. The Temuan tribe were significantly less likely (OR = 0.444, 95%CI= 0.254 – 0.777) to have increment in prevention practices score compared the reference group (other tribe). The findings showed that participants with primary (odds ratio (OR) =2.627. 95% Confidence Interval (Cl): (1.338-5.160, p< 0.05) and secondary level education (OR =2.263. 95% Cl:(1.126- 4.550, p<0.05) were more likely to exhibit improved dengue prevention practices scores. Housewives were less likely to have improved scores of dengue prevention practices (OR=0.535, 95%Cl:(0.289-0.950, p<0.05) after the intervention. Participants with higher dengue knowledge score (7-17) were more likely (OR= 2.390, 95% CI = 1.521 – 3.757, p<0.001) to have higher increment in dengue prevention practices score compared to those scored lower total knowledge score. Participants with no increment in perceived severity (OR= 0.349, 95%CI = 1.521- 3.757, p<0.05) and no increment perceived susceptibility (OR= 0.474, 95%CI = (0.286 – 0.785, p<0.00) were significantly less likely to have higher increment in dengue prevention practices score. Conclusion: Findings imply that the dengue awareness calendar designed for the intervention improved dengue prevention practices score. Thus, this study highlights on the importance of incorporating dengue awareness calendar as one of the implementations in the national program to ensure the positive knowledge and beliefs transform into real dengue prevention practices
A systemic functional linguistic transitivity analysis of character in the Mahabharata, Duryodhana and Draupadi / Shevaanni Pillai
No full textThe current study is a Systemic Functional Linguistic (SFL) (Halliday and Matthiessen,
2014) investigation of character in a famous and timeless Sanskrit epic called The
Mahabharata, a beloved Indian narrative classic and an important piece of history (Doniger,
2020). The data used for the study is an English retold version by Krishna Dharma (1999)
who is a well-known author of Indian classics. Altogether four chapters of the book are used
in relation to the topic, The Dice Game. The current study employs Halliday’s System of
Transitivity as a theoretical framework to analyze two central characters, Duryodhana, a
Kaurava prince and antagonist of the story and Draupadi, a Pandava princess and protagonist
of the story. The methodology will employ Cresswell’s (2017) qualitative textual analysis
and Sriniwass’s (2009) SFL methodology to analyze the data in a systematic way. A total of
620 sentences were analysed regarding the two characters. For Duryodhana, the predominant
Participant role is Actor (15.86%) which has the most frequent characteristics of wickedness,
hot-temperedness and enviousness foregrounding him as a dominant villain. In contrast, for
Draupadi, the predominant transitivity role is Goal (19.26%) which has the most frequent
characteristics of helplessness, valuableness, and wisdom, rendering her the victim of
Duryodhana’s greed and cruelty. In terms of Circumstances, the Circumstance of Manner
was the most prominent found for both Duryodhana (26.42%) and Draupadi (22.78%) which
showed many adverbial and adjectival phrases showing their characteristics. Similarly, the
Material process (Transformative-Transitive) was most prominent Process type found for
both Duryodhana 26.12% and Draupadi 39.18%. The study has shown that the System of Transitivity was indeed a useful tool to study how characters are constructed through a
linguistic analysis of a narrative text. The significance of the study is that it shows how
characters can be interpreted in an objective way using transitivity patterns found in the text
The use of modal auxiliaries as hedging devices in Chinese research articles / Yang Anqi
No full textThe study uses the modal auxiliaries as the principal strategy to find out its variation in Chinese research articles. The study’s goal is to provide a common view for selecting Chinese modal auxiliaries, which will serve as a reference for researchers as they work to enhance their articles. The research took six
representative main modal auxiliaries of hedges in Chinese as the research
objectives and researched their frequencies and distributions in 40 Chinese
research articles from four different disciplines. Based on the previous studies,
the research proposed and answered four research questions and analyzed the data
in a comparative way. It is found that if the context is as concrete as possible, the
Chinese modal auxiliaries would achieve semantic equivalence with English in
academic discourse. Besides, the article summarized and discussed the
scholars’ preferences about modal auxiliaries used in research articles