• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.2
• /
• pp.9-22
• /
• 2004

This paper presents a parallel genetic algorithm for the Multi-disk data allocation problem an NP-complete problem. This problem is to find a method to distribute a Binary Cartesian Product File on disk-arrays to maximize parallel disk I/O accesses. A Sequential Genetic Algorithm(SGA), DAGA, has been proposed and showed the superiority to the other proposed methods, but it has been observed that DAGA consumes considerably lengthy simulation time. In this paper, a parallel version of DAGA(ParaDAGA) is proposed. The ParaDAGA is a 2-dimension torus-based Parallel Genetic Algorithm(PGA) and it is based on a distributed population structure. The ParaDAGA has been implemented on the parallel computer simulated on a single processor platform. Through the simulation, we study the impact of varying ParaDAGA parameters and compare the quality of solution derived by ParaDAGA and DAGA. Comparing the quality of solutions, ParaDAGA is superior to DAGA in all cases of configurations in less simulation time.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.12 no.3
• /
• pp.451-456
• /
• 2017

Semantic segmentation, which is the most basic and complicated problem in computer vision, classifies each pixel of an image into a specific object and performs a task of specifying a label. MRF and CRF, which have been studied in the past, have been studied as effective methods for improving the accuracy of pixel level labeling. In this paper, we propose a semantic partitioning method that combines CNN, a kind of deep running, which is in the spotlight recently, and CRF, a probabilistic model. For learning and performance verification, Pascal VOC 2012 image database was used and the test was performed using arbitrary images not used for learning. As a result of the study, we showed better partitioning performance than existing semantic partitioning algorithm.

• Aerospace Engineering and Technology
• /
• v.7 no.2
• /
• pp.1-10
• /
• 2008

The Bird strike to small aircraft has not been an issue because of it's low speed and usage as a private aircraft. So, the compliance of the bird strike regulation is limited to large fixed-wing aircraft such as the commuter category in FAR Part 23 and the civil aircraft in FAR Part 25, generally. However, the forecast of dramatic increasing of VLJ(Very Light Jet) and (light time of general aviation due to Air-taxi for the point to point transportation, would rise up the need of bird strike regulations and a safety enhancement in normal and utility categorized aircraft. In this study, the safety enhancement concept using a crushable foam for the bird strike to small aircraft wing leading edge, and the evaluation about the safety of the bird strike to small aircraft are proposed using the explicit finite element analysis.

• Journal of Korea Multimedia Society
• /
• v.3 no.3
• /
• pp.253-262
• /
• 2000

This paper proposes a new RBF neural network that determines the number and the center of hidden neurons based on the adaptive feature extraction for the input data. The principal component analysis is applied for extracting adaptively the features by reducing the dimension of the given input data. It can simultaneously achieve a superior property of both the principal component analysis by mapping input data into set of statistically independent features and the RBF neural networks. The proposed neural networks has been applied to classify the 200 breast cancer databases by 2-class. The simulation results shows that the proposed neural networks has better performances of the learning time and the classification for test data, in comparison with those using the k-means clustering algorithm. And it is affected less than the k-means clustering algorithm by the initial weight setting and the scope of the smoothing factor.

• Journal of Intelligence and Information Systems
• /
• v.7 no.1
• /
• pp.17-25
• /
• 2001

Tree model is the most popular classification algorithm in data mining due to easy interpretation of the result. In CART(Breiman et al., 1984) and C4.5(Quinlan, 1993) which are representative of tree algorithms, the split fur classification proceeds to attain the homogeneous terminal nodes with respect to the composition of levels in target variable. But, fur instance, in the chum prediction modeling fur CRM(Customer Relationship management), the rate of churn is generally very low although we are interested in mining the churners. Thus it is difficult to get accurate prediction modes using tree model based on the traditional split rule, such as mini or deviance. Buja and Lee(1999) introduced a new split rule, one-sided purity for classifying minor interesting group. In this paper, we compared one-sided purity with traditional split rule, deviance analyzing churning vs. non-churning data of ISP company. Also reviewing the result of tree model based on one-sided purity with some simulated data, we discussed problems and researchable topics.

• Journal of Korean Ophthalmic Optics Society
• /
• v.9 no.2
• /
• pp.301-312
• /
• 2004

This study was performed to compare the relationship between contact lens wearer and dry-eye patient. In this study, TBUT, SIT, TTT, Rose bengal staining and McMonnies dry eye symptom questionnaire were performed as a baseline. With the base data, the subjects were classified to 3 groups : 3S patients who have dry eye signs and symptoms, 38 patients who are wearing soft contact tenses, 35 subjects who have health eyes and never worn on a contact lenses as control subjects. Contact lens wearers were divided into 3 groups according to the duration of contact lens wear. There were no significant differences in TBUT, STT, TTT, Rose-bengal staining and McMonnies dry eye symptom questionnaire result between contact lens wearer group and dry eye patients group. We suggest that there are similarities in tear function tests between the dry-eye patient and the contact lens wearer.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.04a
• /
• pp.53-56
• /
• 2000

The problem ( Ρ ) of optimizing a linear function d$\^$T/x over the set of efficient set for a multiple objective linear program ( Μ ) is difficult because the efficient set is nonconvex. There some interesting properties between the objective linear vector ｄ and the matrix of multiple objectives C and those properties lead us to establish criteria to solve ( Ρ ) with a linear program. In this paper we investigate a system of the linear equations C$\^$T/${\alpha}$=d and construct two linearly independent positive vectors ${\mu}$, ν such that ${\alpha}$=${\mu}$-ν. From those vectors ${\mu}$, ν, solving an weighted sum linear program for finding an efficient extreme point for the ( Μ ) is a way to get an optimal solution ( Ρ ). Therefore our theory gives an easy way of solving nonconvex program ( Ρ ) with a weighted sum linear program.

• Journal of Broadcast Engineering
• /
• v.6 no.2
• /
• pp.177-186
• /
• 2001

A digital watermark Is a perceptually unobtrusive signal embedded in some multimedia asset such as an Image for copyright protection. In many cases watermark detection amounts to thresholding a correlation vague between a watermark and a received image. Watermarking detection schemes can be classified into two types. Type 1 is based on a correlation process that is applied to the difference between an original image and an input Image to be tested. Type 2 is based on a correlation process that is directly applied to an input Image. The type 1 fails to prove the rightful ownership, while type 2 has an advantage in terms of rightful ownership compared with type 1. However, type 2 has a problem that doesnt appear in type 1. The problem is that correlation between a watermark and an original Image to be watermarked is trio significant to be ignored, when it Is normalized by watermarks energy. In this paper, based on the analysis of the correlation, we propose a novel watermarking scheme to minimize the effect and also verify the performance of the proposed scheme by experiments.

• The Proceedings of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.10 no.2
• /
• pp.63-70
• /
• 1996

In this paper, we propose the low sldelobe digital doppler filter bank design algorithm which is practically applicable to coherent radar equipment in complex air-defense system. Using the proposed algorithm, we can obtain a digital doppler filter bank of lower peak sidelobe level and higher clutter suppression capability. In this method, it is possible to achieve higher target-detectability without loss of Target SNR. The proposed algorithms are verified by simulations and experiments.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.8
• /
• pp.372-377
• /
• 2008

Multiclass cancer classification has been actively investigated based on gene expression profiles, where it determines the type of cancer by analyzing the large amount of gene expression data collected by the DNA microarray technology. Since gene expression data include many genes not related to a target cancer, it is required to select informative genes in order to obtain highly accurate classification. Conventional rank-based gene selection methods often use ideal marker genes basically devised for binary classification, so it is difficult to directly apply them to multiclass classification. In this paper, we propose a novel method for multiclass gene selection, which does not use ideal marker genes but directly analyzes the distribution of gene expression. It measures the class-discriminability by discretizing gene expression levels into several regions and analyzing the frequency of training samples for each region, and then classifies samples by using the naive Bayes classifier. We have demonstrated the usefulness of the proposed method for various representative benchmark datasets of multiclass cancer classification.