MMU Press (Multimedia University)
Not a member yet
714 research outputs found
Sort by
Freedom Indices and Capital Asset Pricing Model: Malaysian Evidence Influence Policy? DOI: https://doi.org/10.33093/ijomfa.2020.1.1.10
No full textHuman rights and fundamental freedoms such as economic, political, and press freedoms vary widely from country to country. It creates opportunity and risk in investment decisions. Thus, this study is carried out to examine if the explanatory power of the model for capital asset pricing could be improved when these human rights movement indices are included in the model. The sample for this study comprises of 495 stocks listed in Bursa Malaysia, covering the sampling period from 2003 to 2013. The model applied in this study employed the pooled ordinary least square regression estimation. In addition, the robustness of the model is tested by using firm size as a controlled variable. The findings show that market beta as well as the economic and press freedom indices could explain the cross-sectional stock returns of the Malaysian stock market. By controlling the firm size, it adds marginally to the explanation of the extended CAP model which incorporated economic, political, and press freedom indices
Digital Mapping of UMK Jeli Campus Using Drone Technology
No full textAerial or satellite images are conventionally used for geospatial data collection and in producing a topographic map. The Unmanned Aerial Vehicles (UAV) technology such as drone has developed by providing very high spatial and temporal resolution data at a lower cost. Nowadays, drones not only use for military purpose but also been utilized widely by the public community for mapping, monitoring, video capturing activities and as a hobby. This present study focuses on the utilization of drone technology to produce a digital map of UMK Jeli Campus. The objective of this study is to access the capability and the accuracy of the drone in producing a digital map. Parrot ANAFI and DJI FC6310 devices were used as a platform to acquire digital images of the study area. After capturing the digital images, ground control points were established with the aid of a handheld global positioning system (GPS) device. Images were processed using Agisoft Photoscan software to produce a digital map of UMK Jeli Campus. This study shows that UAV can be used for producing a digital map at sub-meter accuracy and it can also be used for diversified applications
Effect of Dam Construction to Land Use and Land Cover Changes Using Remote Sensing: A Case of Hulu Terengganu Hydroelectric Dam
No full textIncrease demand of electricity has driven the exploration of Hulu Terengganu hydroelectricity dam as an option to generate electric power supply. Therefore, as a way to monitor the landscape changes, spectral indices were used to study on the spatial-temporal changes over the Hulu Terengganu catchment area. Spectral indices technique was applied on satellite images which acquired on three different years to describe the changes before the construction phase, during the construction phase and post-construction phase. Satellite images used in this study are SPOT-5 and Landsat 7 for year 2006, SPOT-5 and Landsat 8 for year 2014 and Spot-7 and Sentinel 2 for year 2018. Alteration of the land use and land cover has recorded degradation of the forest area by -5.53 % during the dam construction phase (2014) and -8.35 % during post-construction phase (2018). It also found that water body has remarkable change of 10,141% increase from 2014 to 2018. The result from this study would be useful as an overview for future planning, decision making and dam management activity related to the Hulu Terengganu hydroelectric dam operation
Spatial Temporal Anisotropic Transport of the Microalgae Chlorella Vulgaris in the Microfluidic Channel
No full textTransport of biomaterials in confined geometry exhibit complex dynamics, both in spatial and temporal scales. In this work, we examined the transport behavior of microalgae Chlorella Vulgaris inside the microfluidic channel under pressure-driven Poiseuille flow environment. The microalgae system is treated as the spherical naturally buoyant particles. The algae particles trajectories are visually traced using particle imaging techniques and the sample paths are resolved in streamwise (flow) and perpendicular directions. In order to understand boundary wall effects on the flow, we partitioned the microfluidic channel into three different regions, namely the center region and two near-wall boundary regions based on the velocity flow profile. Time-averaged mean square displacement (MSD), probability density function (PDF), skewness, and kurtosis of finite ensembles of particle trajectories are determined for these regions. In addition to the spatial dependencies, we also examined the transient characteristics of the algae transport at early-time and long-time. We found the existence of the mixed types for transport dynamics irrespective of flow region separation often assumed in many laminar flow simulations. This finding will be useful for optimization of mixing of algae culture in micro-scale photobioreactor as the productivity of cell growth depends critically on the cell dispersion or transport
VHDL Modelling of Low-Cost Memory Fault Detection Tester
No full textMemory modules are widely used in varies kind of electronics system design. The capacity of the memory modules has increased rapidly since the past few years in order to satisfy the high demand from the end-users. The memory modules’ manufacturers demand more units of automatic test equipment (ATE) to increase the production rate. However, the existing ATE used in the industry to carry out the memory testing is too costly (at least a million dollars per ATE tester). The low-cost memory testers are urgently needed to increase the production rate of the memory module. This has inspired us to design a low-cost memory tester. A low-cost memory fault detection tester with all the major fault detection algorithms that used in industry is modelled using Very High Speed Integrated Circuit Hardware Description Language (VHDL) in this paper to support the need of the low-cost ATE memory tester. The fault detection algorithms modelled are MATS+ (Modified Algorithm Test Sequence), MATS++, March C, March C-, March X, March Y, zero-one and checkerboard scan tests. PERL program is used to analyse the simulation results and a log file will be generated at the end of the memory test. Extensive simulation and experimental test results show that the memory tester modelled covers all the memory test algorithms used in the industry. The low-cost memory fault detection tester designed provides the 100 % fault detection coverage for all memory defects
An FPGA based Real-Time Multi-Target Synthetic Aperture Radar Echoes Synthesizer
No full textThis paper proposes a technique for synthesizing multiple point target scatterer Synthetic Aperture Radar (SAR) echoes in real-time. Traditional approaches require high computation resources to calculate the complex SAR echoes due to its complex mathematical model. The proposed technique employs the low computation Direct Digital Synthesis (DDS) approach to generate these complex sinusoid echoes. The proposed Synthetic Aperture Radar Echoes Synthesizer (SAR-ES) is capable of synthesizing SAR echoes accurately in real-time and was built in a Field Programmable Gate Array (FPGA) platform. The system can be used as a testbed to validate and evaluate the performance of a real-time SAR processing algorithms/system prior to the actual flight mission. This could help in reducing the frequency of flight trials and to reduce the SAR system development risk especially for satellite-borne SAR system
The Nexus of Female Labour Force Participation, Economic Growth, Education and Fertility Rate: Empirical Evidence in Malaysia: DOI: https://doi.org/10.33093/ijomfa.2020.1.1.2
No full textMalaysia, a fast-growing developing country in Asia, has envisioned Shared Prosperity Vision 2030 to become a developed economy with highincome via sustainable and inclusive economic growth by the year 2030. To accomplish this vision, femalelabour participation isneeded as the femalepopulation constitutesalmost half of Malaysia’s total population. However, female labour participation rate iswaylower than Malaysia’s overall labour force participation rate.The relatively low female labour participation rate can be a barrier to Malaysia’s economic development and thus the realization of its goal of a high income nation.Therefore, this paper makes an attempt to examine empirically the long-run causal association amongfemale labour force participation, economic growth, education, and fertility rate. The interrelationships among the variables are examined using the bounds test and Toda-Yamamoto granger non-causality methodology. The result of the study indicates a strong evidence of long-run relationship among the variables. Besides, we have found asignificant inverted-U-shaped associationlinking the female labour force participationto the economic growthin Malaysia. The results of Granger causality tests further confirm that there is a strong evidence of bidirectional causality from education to economic growth as well as female labour participation. Besides, the results also show significant unidirectional causality from female labour force participation and fertility to economic growth
I-Bin: Weight Based IoT Smart Recycling Scheduler for Guarded Neighbourhood
No full textOverpopulation and lack of awareness are the main causes of poor waste management. While existing research in waste management employs current technology such as the Internet of Things, it lacks emphasis on residential centric type of waste management system. This project designs a weight based scheduling system for a guarded neighbourhood using Arduino Uno, load cells and plastic bins which are then incorporated with WiFi to send waste weight information in real time to cloud for monitoring. Design verification tests such as the linearity test and non- repeatability test showed less than 1 percent standard deviation error. A Proof of Concept test was conducted to test the system's performance at a guarded residential area. Analysis of test showed that an average of 0.0966 kg of recyclable waste was collected per house. Based on the results also, it is predicted that approximately 483 kg of waste can be effectively collected from 10 residential areas using the I-Bin system. Residents tend to dispose waste after office hours and scheduling more waste collection frequency after office hours will lead to increased revenue for the recycling company.
Study of Temperature Measurement Accuracy by Using Different Mounting Adhesives
No full textThermal compounds are adhesive used to improve heat conduction between two surfaces. It can be used to secure a thermocouple to a surface which the temperature is being measured. This paper studies the temperature accuracy when using different types of thermal adhesives to secure thermocouples to a metal surface. An aluminum block attached to heater resistors was heated up by supplying varying power levels to create different temperatures. The measured temperature is compared to a reference thermocouple in the aluminum block to check the accuracy of each thermocouple when it is secured with different adhesives. It was found using the Loctite 3873 to secure a thermocouple to a metal surface will produce the most accurate temperature reading with an error below 2.6oC. This enables researchers to use the appropriate adhesive to obtain the most accurate results and also to know what are the errors contributed by different adhesives.
Convolutional Neural Network-based Transfer Learning and Classification of Visual Contents for Film Censorship
No full textContent filtering is gaining popularity due to easy exposure of explicit visual contents to the public. Excessive exposure of inappropriate visual contents can cause devastating effects such as the growth of improper mindset and rise of societal issues such as free sex, child abandonment and rape cases. At present, most of the broadcasting media sites are hiring censorship editors to label graphic contents manually. Nevertheless, the efficiency is limited by factors such as the attention span of humans and the training required for the editors. This paper proposes to study the effect of usage of Convolutional Neural Network (CNN) as feature extractor coupled with Support Vector Machine (SVM) as classifier in an automated pornographic detection system. Three CNN architectures: MobileNet, Visual Geometry Group-19 (VGG-19) and Residual Network-50 Version 2 (ResNet50_V2), and two classifiers: CNN and SVM were utilized to explore the combination that produce the best result. Frames of films fed as input into the CNN were classified into two groups: porn or non-porn. The best accuracy was 92.80 % obtained using fine-tuned ResNet50_V2 as feature extractor and SVM as classifier. Transfer learning and SVM have improved the CNN model by approximately 10 %