Articles
9,199 Documents
An Efficient Content Based Image Retrieval Scheme
Zukuan WEI;
Hongyeon KIM;
Youngkyun KIM;
Jaehong KIM
Indonesian Journal of Electrical Engineering and Computer Science Vol 11, No 11: November 2013
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Due to the recent improvements in digital photography and storage capacity, storing large amounts of images has been made possible. Consequently efficient means to retrieve images matching a user’s query are needed. In this paper, we propose a framework based on a bipartite graph model (BGM) for semantic image retrieval. BGM is a scalable data structure that aids semantic indexing in an efficient manner, and it can also be incrementally updated. Firstly, all the images are segmented into several regions with image segmentation algorithm, pre-trained SVMs are used to annotate each region, and final label is obtained by merging all the region labels. Then we use the set of images and the set of region labels to build a bipartite graph. When a query is given, a query node, initially containing a fixed number of labels, is created to attach to the bipartite graph. The node then distributes the labels based on the edge weight between the node and its neighbors. Image nodes receiving the most labels represent the most relevant images. Experimental results demonstrate that our proposed technique is promising.DOI: http://dx.doi.org/10.11591/telkomnika.v11i11.3588
Equipment maintenance support capability evaluation using cloud barycenter evaluation method
Hongqiang Gu;
Cheng Zhang;
Quan Shi
Indonesian Journal of Electrical Engineering and Computer Science Vol 11, No 2: February 2013
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Maintenance support is the most important measures to keep equipments having high operational capabilities. Equipment maintenance support capability is an important part of operational capability and the evaluation of equipment maintenance support capability is very important to the establishment of battle effectiveness. The evaluation index system of equipment maintenance support capability is established according to the evaluation index establishing principles. Cloud barycenter evaluation method is applied to equipment maintenance support capability evaluation on basis of the established evaluation index system. The application steps of cloud barycenter evaluation method to equipment maintenance support capability evaluation are analyzed. A calculating example for equipment maintenance support capability using the proposed algorithm is presented and the evaluation results are achieved using the weighted deflection degree which is used to demonstrate the deflection degree between equipment maintenance support capability and its perfect state. The correctness and validity of the proposed method is verified by the calculating results, which provide an efficient method for equipment maintenance support capability evaluation. DOI: http://dx.doi.org/10.11591/telkomnika.v11i2.1982
Research of Plant-Leaves Classification Algorithm Based on Supervised LLE
Yan Qing;
Liang Dong;
Zhang Jingjing
Indonesian Journal of Electrical Engineering and Computer Science Vol 11, No 6: June 2013
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
A new supervised LLE method based on the fisher projection was proposed in this paper, and combined it with a new classification algorithm based on manifold learning to realize the recognition of the plant leaves. Firstly,the method utilizes the Fisher projection distance to replace the sample's geodesic distance, and a new supervised LLE algorithm is obtained .Then, a classification algorithm which uses the manifold reconstruction error to distinguish the sample classification directly is adopted. This algorithm can utilize the category information better,and improve recognition rate effectively. At the same time, it has the advantage of the easily parameter estimation. The experimental results based on the real-world plant leaf databases shows its average accuracy of recognition was up to 95.17%. DOI: http://dx.doi.org/10.11591/telkomnika.v11i6.2680
Fuzzy-PID Controller of robotic grinding force servo system
Adnan Jabbar Attiya;
Yang Wenyu;
Salam Waley Shneen
Indonesian Journal of Electrical Engineering and Computer Science Vol 15, No 1: July 2015
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
When a robot is used to grind or finish a curved surface ,as marine propeller surface, both contact force and feed movement must controlled at the similar time in order that the grinding tool would machine the work-piece, with required force, at the right position in right posture. A compliant wrist system is advanced, in this paper, to conform the shape of the machining propeller by altering its posture along with the surface. Grinding force is controlled under a simple new Fuzzy-PID controller with five input variables which assembled and compared with an antecedently used PID controller. The aim of defining the rules and its optimization are to achieve a controller that provides grinding with higher quality. Both the controllers PID and Fuzzy-PID have been optimized together with the parameters of the Two-Phase Hybrid Stepping Motor The Fuzzy-PID controller policy at a steady value in the normal direction of the mentioned machining point by multi-point machining, while the grinding tool moving along the curved surface of the propeller. It means that the model of the compliant wrist system and the surroundings could be used in force controlling when robots grind marine propeller surface by a grinding tool with multi-point machining. DOI: http://dx.doi.org/10.11591/telkomnika.v15i1.8051
The strategy of improving convergence of genetic algorithm
Jiang Jing;
Meng Lidong
Indonesian Journal of Electrical Engineering and Computer Science Vol 10, No 8: December 2012
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Premature convergence is the main obstacle to the application of genetic algorithm. The study on convergence of GA is always one of the most important theoretical issues. Via analyzing the convergence rate of GA, the average computational complexity can be implied and the optimization efficiency of GA can be judged. This paper proposed an approach to calculating the first expected hitting time and analyzed the bounds of the first hitting time of concrete GA using the proposed approach. And this paper proposed a strategy which included transformation of fitness function, self-adaptive crossover and mutation probability and close relative breeding avoidance method in order to overcome premature convergence. DOI: http://dx.doi.org/10.11591/telkomnika.v10i8.1641
The Application of Compressive Sensing on Spectra De-noising
Mingxia Xiao;
Lu Changhua;
Ma Xing;
Jiang Weiwei
Indonesian Journal of Electrical Engineering and Computer Science Vol 11, No 10: October 2013
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Through the analyzing of limitations on wavelet threshold filter de-noising, this paper applies wavelet filter based on compressed sensing to reduce the signal noise of spectral signals, and compares the two methods through experiments. The results of experiments shown that the wavelet filter based on compressed sensing can effectively reduce the signal noise of spectral signal. The de-noising effect of the method is better than that of wavelet filter. The method provides a new approach for reducing the signal noise of spectral signals. DOI: http://dx.doi.org/10.11591/telkomnika.v11i10.3198
Big Data Platforms and Techniques
Salisu Musa Borodo;
Siti Mariyam Shamsuddin;
Shafaatunnur Hasan
Indonesian Journal of Electrical Engineering and Computer Science Vol 1, No 1: January 2016
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijeecs.v1.i1.pp191-200
Data is growing at unprecedented rate and has led to huge volume generated; the data sources include mobile, internet and sensors. This voluminous data is generated and updated at high velocity by batch and streaming platforms. This data is also varied along structured and unstructured types. This volume, velocity and variety of data led to the term big data. Big data has been premised to contain untapped knowledge, its exploration and exploitation is termed big data analytics. This literature reviewed platforms such as batch processing, real time processing and interactive analytics used in big data environments. Techniques used for big data are machine learning, Data Mining, Neural Network and Deep Learning. There are big data architecture offerings from Microsoft, IBM and National Institute of Standards and Technology. Big data potentials can transform economies and reduce running cost of institutions. Big data has challenges such as storage, computation, security and privacy
A novel balanced scorecard design based on fuzzy analytic network process and its application
Yuhong Cao;
You Jianxin
Indonesian Journal of Electrical Engineering and Computer Science Vol 12, No 4: April 2014
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
In this paper, wepropose a novel balanced scorecard design based on fuzzy analytic networkprocess and then conduct performance evaluation through a case study. Afteranalyzing the related works about balanced scorecard design and the algorithmof fuzzy analytic network process, we illustrate the improved balancedscorecard design. Firstly, the basic concepts for the balanced scorecardsare introduced. Secondly, four perspectives of the balanced scorecards design are provided.Thirdly, the method to promote the performance of balanced scorecard designthrough the fuzzy analytic network process is demonstrated. In the proposeddesign, a fuzzy number is represented by the left and right formation of eachdegree of membership in fuzzy analytic network process. Furthermore, the degreepossibility for a convex fuzzy number to be larger than a given convex fuzzynumbers can be represented by an effective scheme. Particularly, the basicstructure in the fuzzy analytic network process model is organized hierarchically,and the local weights of the strategies, balanced scorecard perspectives andperformance indicators can be obtained by matrix computing. Finally, a casestudy of college English classroom teaching quantitative evaluation is given todemonstrate the performance of the proposed balanced scorecard design.Experimental results show that the proposed balanced scorecard design is quiteeffective. DOI : http://dx.doi.org/10.11591/telkomnika.v12i4.4804
Predictive Function Control for Milling Process
Zhihuan Zhang;
Chao Hu
Indonesian Journal of Electrical Engineering and Computer Science Vol 11, No 8: August 2013
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
The adaptive constant control for the cutting process is an effective way to improve the productivity of the Computer Numerical Control (CNC) milling machine. This control is achieved through adjusting the feed rate online, and many scholars has been concentrated in the field. However, in of the existing adaptive constant control algorithms, the controller parameters are tuned only depends on the dynamic behavior of the controlled system, without giving effect to control the input and system output prospects. Therefore, milling force mutated because of the mutation of the depth or width of cut usually resulting in the overshooting of control system or overflow of control input. In this paper, we present a new solution to the problem, in which a mathematical model of milling process is established based on the characteristics of the CNC milling process. Then the predictive functional control law is introduced on the milling process and the controller parameters can be tuned online. The Simulation results show that the proposed method has the advantages of strong robustness to different interferences, good practicability for the milling process, and good real-time control responsibility. DOI: http://dx.doi.org/10.11591/telkomnika.v11i8.3042
Simulation of dual stage thulium-doped fiber amplifier using pump power distribution technique
Muhammad Syauqi Kusyairi bin Jamalus;
Nelidya Md. Yusoff;
Abdul Hadi Sulaiman
Indonesian Journal of Electrical Engineering and Computer Science Vol 15, No 3: September 2019
Publisher : Institute of Advanced Engineering and Science
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.11591/ijeecs.v15.i3.pp1203-1211
This paper shows dual stage thulium-doped fiber amplifiers (TDFAs) that use a pump power distribution technique. Simulations were done with signals ranging from 1975 nm to 2000 nm using the OptiSystem v.13 software. The results required were gathered from the software. The results of gain, noise figure, optical signal-to-noise ratio (OSNR) and output power were obtained. The highest gain and lowest noise figure results were achieved for the double pass dual stage TDFA configuration with values of 19.85 dB and 5.58 dB respectively, followed by the single pass dual stage TDFA. The OSNR and output power performances were also better for the double pass dual stage TDFA, obtaining 57.12 dB and 19.55 dBm respectively. This study shows that thulium can be used in the 2 µm region as an active gain medium and the dual stage architecture and distributed pumping technique proves to be effective techniques to obtain the desired results. Experimental work will be done in the future with the simulated results used as a reference.