Proceeding of the Electrical Engineering Computer Science and Informatics
Not a member yet
649 research outputs found
Sort by
Pattern Recognition Approach for Formation Control for Swarm Robotics Using Fuzzy-Kohonen Networks
Get PDFRecent advances in robotics have started makingit feasible to deploy large numbers of inexpensive robotsfor tasks such as surveillance and search. However,coordination of multiple robots to accomplish such tasksremains a challenging problem. This paper describes newalgorithm Fuzzy-Kohonen Networks (FKN) to achieve realtimereactive formation control for swarm robots
Enhanced Ridge Direction for the Estimation of Fingerprint Orientation Fields
Get PDFAn accurate estimation of fingerprint orientation fields is an important step in the fingerprint classification process. Gradient-based approaches are often used for estimating orientation fields of ridge structures but this method is susceptible to noise. Enhancement of ridge direction improves the structure of orientation fields and increases the number of correct features thereby conducing the overall performance of the classification process. In this paper, we propose an algorithm to improve orientation field structures using variance of gradient. That algorithm have two steps; firstly, estimation of fingerprint orientation fields using gradient-based method, and finally, enhancement of ridge direction using minimum variance of the cross center block direction. We have used standard fingerprint database NIST-DB14 for testing of proposed algorithm to verify the degree of efficiency of algorithm. The experiment results suggest that our enhanced algorithm achieves visibly better noise resistance with other methods
Building Student’s Study Path using Markov Chain Process with Apriori Cross Join Pearson Correlation
Get PDFStudent’s study path could be advised by using bestpossible path from Markov Chain rule based on student’sacademic performance records with several assumption on thecurrent curriculum. Finding the Markov’s rule is crucial processbecause it will determine study path’s scenarios which rely onstudent current performance to choose the next best possiblepath. The rule would be built using the whole student’s academicperformance on the same curriculum by implementing AprioriCross Join Pearson Correlation Test on two consecutivesemesters. It will then create path consist of paired courses A->B with Pearson value that would be implemented as rule in Markov Proces
Car Engine Storage and Spare Parts Management System Using RFID Technology
Get PDFvery year, demanding of spare parts for vehicle isincreasing together to the number of cars, thus spare parts toreplacing faulty component also high demand. Every workshopneed to keep spare parts for replacing breakdown vehicle or someworkshop request spare parts to store it. Base on that, a good spareparts management and storage system is required for effective andfaster service to thecustomer. This paper discuss on the use ofRadio Freqeuncy Identification (RFID) for car engine storage andspare parts management system using RFID technology, because ofits advantages compare to others technology which is no line ofsight required and multiple reading. Every spare parts or enginetagging by an RFID tag which is programmed early with uniqueidentity, then store to a central database for stock and inventorysystem. Testing havebeen conducted an shows reduce on reading performance becauseof most spare parts and engine made from metal material thuseffected to RFID reading performance. A new technique used forRFID tag before tagging which is make some gap from the engineor spare parts to avoid RFID signal absorve by metal material. Theother way is used high gain antenna andRFID reader for faster and powerfull reading. Results showsimprovement on reading distance and performance compare to theprevious tagging technique and tagging placement
Computer Vision Based Object Tracking as a Teaching Aid for High School Physics Experiments
Get PDFExperiments play a vital role in science education. In high school physics, especially in mechanics, many experiments are conducted where tracking a single or multiple objects are required. In most situations students visually observe the motion of objects and take the measurements. This manual method is time consuming, generates higher error and incapable of producing multiple readings rapidly. The research described in this work introduces a simple mechanism to integrate computer vision based tracking to enhance the quality of measurements and to new ways of looking at experiments. The case study consists of three standard experiments. In the first experiment a motion of the simple pendulum was tracked. Using computer vision students were able to obtain a correlation of 0.99 between the calculated period and the theoretical period. In addition, it was possible to calculate the position and the velocity of the bob more than 30 times during a single oscillation. Students were able to plot the extra data points for a better understanding of the simple harmonic motion, which was not possible in the manual method. Second experiment was focused on measuring the terminal velocity of a ball moving through a viscous medium. Final case study was on tracking multiple particles in a moving fluid. In all three experiments computer vision based system provided more accurate and higher number of data points than the manual method. This helps students to understanding the underline theory better. The tracking system was consisted of a digital camera, image preprocessing sub system, feature extraction subsystem, object identification subsystem and data export subsystem. The system was successfully tested on a normal PC which is cost effective to be used in high schools. Based on the case studies it was concluded that such systems can be used in high schools to improve the quality of experiments conducted.
Ontology-Based Sentence Extraction for Answering Why-Question
Get PDFMost studies on why-question answering system usually used the keyword-based approaches. They rarely involved domain ontology in capturing the semantic of the document contents, especially in detecting the presence of the causal relations. Consequently, the word mismatch problem usually occurs and the system often retrieves not relevant answers. For solving this problem, we propose an answer extraction method by involving the semantic similarity measure, with selective causality detection. The selective causality detection is applied because not all sentences belonging to an answer contain causality. Moreover, the motivation of the use of semantic similarity measure in scoring function is to get more moderate results about the presence of the semantic annotations in a sentence, instead of 0/1. The semantic similarity measure employed is based on the shortest path and the maximum depth of the ontology graph. The evaluation is conducted by comparing the proposed method against the comparable ontology-based methods, i.e., the sentence extraction with Monge-Elkan with 0/1 internal similarity function. The proposed method shows the improvements in term of MRR (16%, 0.79-0.68), P@1 (15%, 0.76-0.66), P@5 (14%, 0.8-0.7), and Recall (19%, 0.86-0.72)
Region of Interest Detection for Pregnancy Image Processing
Get PDFThis study discussed on a comparison of three cropping techniques for region of interest (ROI) detection of ultrasonography (USG) image. Ultrasound images are used to provide information about fetal development in the womb. The image generated by the two-dimensional ultrasound has not been able to provide complete information. Therefore, in order get the form of fetus on ultrasound image can be clearly identified with the necessary process of image analysis that can detect the boundaries of objects ROI, so that it can differentiate between one object with another object on the ultrasound image. Comparison results between the ROI detection with cropping methods namely rectanguler, circle and ellipse shapes are expected obtaining the best technique. It can be used in the ultrasound image segmentation process to obtain the best shape of the object of USG image. Based on the available data it can be concluded that the best ROI detection with cropping methods is produced by ellipse shape. The results show that ellipse shape of ROI Detection has the best accuracy compared to other cropping shapes. Based on the results, rectanguler and circle have the similar value. Therefore, MSE of ellipse is lower than rectanguler and circle shapes.
PID Designs Using DE and PSO Algorithms for Damping Oscillations in a DC Motor Speed
Get PDFIn this paper, a design and a simulation of a PID controller are presented to control a DC motor speed. The design is proposed to increase a dynamic stability of the motor speed. Changing the speed setpoint values may generate oscillations, thus the controller is required to maintain a system stability. The PID controller is tuned using Differential Evolutionary (DE) and Particle Swarm Optimization (PSO) algorithms. Performance of both algorithms is compared by Integral Square Error (ISE) and rise time values. Experiments are performed by changing the speed setpoint values. In simulation results, the ISE and the rise time values of the DE algorithm are slower than the PSO algorithm
Measurement of Partial Discharge Induced Electromagnetic Wave using Loop Antenna
Get PDFPartial discharge (PD) detection by detecting the released energy in the form of electromagnetic (EM) wave during discharge using Ultra-High Frequency (UHF) antenna is one of the methods to determine insulation system quality of power apparatus. The main advantage of UHF PD detection system by using antenna are the capability to detect PD signal without firstly shut down the power apparatus. Therefore, the PD measurement system based on EM wave induced by PD detected by the sensor is needed. First, the sensors which be able to detect PD induced EM wave in ultra wide band need to be designed and manufactured. Second, the characteristics of PD induced EM wave measured by the sensors need to be investigated. This paper deals with measurement of PD induced EM wave by using loop antenna. The results show that the further the distance of loop antenna to PD source is, the lower Vpp is. This is due to EM wave attenuation. At the distance 225 cm from PD source, Vpp of PD is equal to Vpp of the background noise (BGN). Therefore, the maximum distance of the loop sensor to PD source is 225 cm
The Effect of Coating on Leakage Current Characteristic of Coast Field Aged Ceramic Insulator
Get PDFCeramic insulator is widely used in Indonesian electrical power distribution system. Based on previous research, it is obtained that silicone rubber coating can improve the performance of ceramic insulator due to better hydrophobicity. This paper presents performance of ceramic insulator with and without silicone rubber coating in a coastal area of Pangandaran. 6 silicone rubber coated and 6 uncoated of 20 kV distribution insulators have been installed in the field to get natural aging process for 5 years and 4 months. Along the aging process, their surface properties have been checked periodically. Monitored parameters are magnitude of leakage current, THD, hydrophobicity, and surface temperature of insulator. The result has shown that after 5 years and 4 months aging in coastal area without maintenance, silicone rubber coated ceramic insulator has better performance