Universiti Teknikal Malaysia Melaka: UTeM Open Journal System
Not a member yet
2735 research outputs found
Sort by
Real-Time Kinect Fingers Tracking and Recognition for Servo Gripper Controller
Get PDFThe conventional method to control something like machines such as remote controllers or wearable sensors have its limitation and cannot cater the high demands in some scenarios. To overcome this situation, a real-time Kinect finger tracking and recognition is developed to control a servo gripper via a single board microcontroller, Arduino. This paper presents hand and finger tracking method that can determine the number of fingers and also the angle of the finger’s position. The hand gesture is captured by the Kinect and go through a series of image processing and finally, the information will send to the Arduino. The image processing method includes a detection of hand using depth sensor on the Kinect, then finds the finger and calculates the angle of the finger’s position. The convex hull algorithm is used to represent the region of the hand. The fingertips are recognized by calculating the angle of the fingertip and each angle is compared with threshold angle. The orientation and position of the fingers are obtained by finding the middle line of the finger and compares to the vertical line of the middle finger, then calculates the angle. The result shows that the number of fingers appeared in the display can be recognized and the orientation and position of each finger can be determined. A gripper also able to react simultaneously (open and close) based on the detected finger
Multi-Task Learning Approach for Natural Images' Quality Assessment
Get PDFBlind image quality assessment (BIQA) is a method to predict the quality of a natural image without the presence of a reference image. Current BIQA models typically learn their prediction separately for different image distortions, ignoring the relationship between the learning tasks. As a result, a BIQA model may has great prediction performance for natural images affected by one particular type of distortion but is less effective when tested on others. In this paper, we propose to address this limitation by training our BIQA model simultaneously under different distortion conditions using multi-task learning (MTL) technique. Given a set of training images, our Multi-Task Learning based Image Quality assessment (MTL-IQ) model first extracts spatial domain BIQA features. The features are then used as an input to a trace-norm regularisation based MTL framework to learn prediction models for different distortion classes simultaneously. For a test image of a known distortion, MTL-IQ selects a specific trained model to predict the image’s quality score. For a test image of an unknown distortion, MTLIQ first estimates the amount of each distortion present in the image using a support vector classifier. The probability estimates are then used to weigh the image prediction scores from different trained models. The weighted scores are then pooled to obtain the final image quality score. Experimental results on standard image quality assessment (IQA) databases show that MTL-IQ is highly correlated with human perceptual measures of image quality. It also obtained higher prediction performance in both overall and individual distortion cases compared to current BIQA models
A Review of Techniques in Automatic Programming Assessment for Practical Skill Test
Get PDFComputer programming ability is a challenging competency that requires several cognitive skills and extensive practice. The increased number of students enrolled in computer and engineering courses and also the increased of failure and drop rate in programming subject is the motivational factor to this research. Due to the importance of this skill, this paper intends to study the landscape of current scenario in assisted assessment for hands-on practical programming focusing on competency-based assessment. The Bloom Taxonomy is used as a competency-based assessment platform. The review showed to-date that there are several automatic assessments for programming skills. However, there is no common grading being applied. Thus, further research is required to propose an automatic assessment that grades the student achievement based on learning taxonomy such as Bloom Cognitive Competency model
Voltage Generation and Thermal Gradient Analysis of a Car Waste Heat using Thermoelectric Generator
Get PDFThis study is conducted to analyse the behaviour of voltage generation and thermal gradient of a car waste heat when these conditions are applied; with no air conditioning system and with air conditioning system. The analysis of this behaviour is made possible by utilising a set of three cascaded thermoelectric generators (TEGs) in each condition. The harvested voltage is attained by placing the hot side of the TEG on top of the dashboard car. A series of data setup from a set of TEGs, temperature data logger and the data acquisition card is accomplished towards the end of the experiment. From the final analysis, it is found that the highest thermal gradient and voltage generation is obtained with the presence of air conditioning
Data Redundancy Reduction in Wireless Sensor Network
Get PDFIn Wireless Sensor Network, sensor nodes are randomly deployed where the sensor nodes are not situated faraway from each other. Thus, an overlapping area is generated due to intersection of their sensing ranges. If an event occurs within the overlapping area, all the sharing nodes sense the same event and produce redundant and correlated data. Data redundancy exhaust network resources and increase network overhead. Data aggregation and numerous data redundancy reduction algorithms are employed to solve this problem. This paper reviews modern data redundancy reduction used sleep schedule model to solve the redundancy. All proposed algorithms are classified on the basis of network coverage and similarity among sensory data which can be used in reducing redundancy in WSN effectively
Persuasive Multimedia in Truancy Awareness (PMTA): Integration of Persuasive Design Principles
Get PDFThis article reports on the ongoing research which focuses on the use of persuasive design principle in the design and development of multimedia learning application for truancy awareness (PMTA).Primarily, this paper emphasizes on how persuasive design principle can be employed into a multimedia learning environment in fostering truancy awareness. This paper is presented in the following order; firstly, the current situation regarding truancy issues is explained. Next, the current practices on how to handle truancy issues are highlighted. The third part discusses the theoretical framework and how the principles were applied in designing the information and presentation of PMTA. In conclusion, it is hoped that PMTA would be a beneficial tool for rising truancy awareness and reducing this problem in Malaysia. The next step of this study will involve with experimental testing to evaluate the effects of the persuasive multimedia applications toward truancy awareness
A Reconfigurable Ultra-Wideband (UWB) Antenna with Single Band-Rejection
Get PDFIn this paper, a reconfigurable ultra-wideband (UWB) antenna with WLAN band-rejection is presented. The proposed antenna operates in the ultra-wideband frequency band from 2.99 – 10.82 GHz with band- rejection of 5.00 – 5.85 GHz. The structure of the antenna is simple octagon-shaped radiating patch and half ground plane with dimension of 30 x 40 mm2. Bandstop elements (inverted U-shaped slot) are integrated into the structure of the antenna to achieve WLAN bandrejection. Band-rejection reconfigurable capability are achieved by placing RF switches at the slot. By configuring the switch in the ON and OFF modes, two reconfigurable states are realized which are WLAN band-rejection during ON states and full UWB frequency spectrum during off states. The proposed antenna shows a wide impedance bandwidth with sharp rejection at the WLAN band and provide multi-function antenna due to its reconfigurable characteristics. The performance was analyzed in terms of reflection loss, VSWR, gain and radiation pattern. Simulation and measurement result have achieved a good agreement
Gas Sensing Mobile Robot: A Review
Get PDFMobile robot applications are required in various hazardous fields to reduce human casualties. One of the most demanding applications is gas sensing mobile robot. Since the hazardous chemical compound is undetectable by humans, autonomous mobile gas sensors are needed. Over the past few decades, various attempts to incorporate gas sensor on mobile robots are reported. Gas source localisation and gas distribution mapping are the two mainly focused scope of research. This paper presents the earliest works and recent development in gas sensing mobile robots
Highly Responsive CO2 Detection by an Improved & Compact Gas Sensor Using Mid-IR Spectroscopy
Get PDFAn improved and compact gas sensor using midinfrared spectroscopy for highly responsive Carbon Dioxide (CO2) detection is presented. The sensing principle is based on open-path direct absorption spectroscopy in the mid-infrared region. The improved reflective structure of optical gas sensor consists of low cost and compact components. Several gas cell configuration which includes SISO, 2-MISO, 4-MISO and 8- MISO were simulated using ZEMAX®12 using non-sequential ray tracing technique to get the optimum radius of the reflective curved surface. Sensitivity analysis has shown that the optimised structure of 4-MISO yields the highest sensitivity and power efficiency of -0.2895%-1 and 12.298% respectively. The developed gas sensing system using the optimised gas cell has shown the capability of accurately detecting CO2 concentration between 1.5% and 5.8% with no cross-sensitivity with other gases present in the gas cell. The newly developed CO2 sensor exhibits high responsivity with the recorded rise time and fall time of less than 1 second respectively
A Caching Algorithm for Information Centric Network Using Fuzzy Logic
Get PDFThe internet today has evolved from information superhighway to a household necessity that offers more than just information. Nowadays, the internet serves a lot of purpose. It is a tool for not just information but entertainment that offers music, graphics and videos that is available for downloading or streaming. It has also evolved to be a medium of communication that offers a global link from people around the globe. From emails, short message services and even voice communication, the internet has all of these to offer. The former information superhighway is today a social media platform that is open to all ages to all variety of users. With this development, it is logical to think that the current internet network scheme should also be subjected to evolution. The emerging Information Centric Network is quite a good fit to the future of internet. The idea to be concerned to the content that is to be accessed more than the identity of the one accessing the content is tailor-fit to the current application of internet. In a nutshell, ICN requires node with caching functionality. An effective caching algorithm is a great help to attain the very purpose of ICN which is to come up with an efficient network. Meanwhile, fuzzy logic, which has proven to be effective in control or optimization applications, can also be applied in improving caching functionality of ICN. This paper explores the application of fuzzy logic to the caching algorithm that can be used to further improve current information centric networks. The results were obtained from hypothetical data because this is just to prove that fuzzy logic can be applied in the caching dynamics of Information Centric Network