• Journal of the Korean Society of Clothing and Textiles
• /
• v.31 no.3 s.162
• /
• pp.364-375
• /
• 2007

The propose of this study is to develop the optimal sizing system of ready-to-wear far elementary school boys using a newly invented statistical technique. The body measurements was classified by the method that equalizes the distribution of the subjects using the probability density function, to theoretically systemize a method to determine a size range of ready-to-wear for elementary school boys between 7 to 12 yeiws old. The results were as follows: 1. Height group includes 9 types of heights: 115, 120, 125, 130, 135, 140, 145, 150 and 155. 2. In the case of short children's groups, the variance in bust girth and waist girth is narrow. The people cluster together around the average. The size deviation of ready-to-wear is small. 3. In the case of tall children's groups, the variance in bust girth and waist girth is wide. The people spread widely around the average. The size deviation of ready-to-wear is large. 4. The optimal size system is suggested considering the weight of growth exponent of children according to their respective ages. Clothing companies can selectively choose sizes that meet the target of their brands. 5. It suggests the body sizes chart, which based on their means by the middle size children for each height group, so that clothing companies make use of it.

• The Korean Journal of Applied Statistics
• /
• v.6 no.2
• /
• pp.371-381
• /
• 1993

When the multicollinearity presents in the standard linear regression model, ridge regression might be used to mitigate the effects of collinearity. As the prediction-oriented criterion, the integrated mean sqare error criterion $J_w(k)$ was introduced by Lim, Choi & Park(1980). By noting the equivalent relationship between the $C_k$ criterion and $J_w(k)$ with a special choice of weight function $W(x)$, we propose a more reasonable selection rule of k w.r.t. the $C_k$ criterion than that given in Myers(1986). Next, to find the $\beta(k)$ which behaves reasonably well w.r.t. competing criteria, we adopt the minimax principle in the sense of maximizing the worst relative efficiency of k among competing criteria.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.464-464
• /
• 2012

수문분야에 있어서 빈도해석의 목적은 특정 재현기간에 대한 발생 가능한 수문량의 규모를 파악하는데 있으며, 빈도해석의 정확도는 적합한 확률분포모형의 선택과 매개변수 추정방법에 의존하게 된다. 일반적으로 각 확률분포모형의 특성을 대표하는 매개변수를 추정하기 위해서는 모멘트 방법, 확률가중 모멘트 방법, 최대우도법 등을 이용하게 된다. 모멘트 방법에 의한 매개변수 추정은 해를 구하기 위한 과정이 단순한 반면, 비대칭형의 왜곡된 분포를 갖는 자료들에 대해서는 부정확한 결과를 나타내게 된다. 확률가중 모멘트 방법은 표본의 크기가 작거나 왜곡된 자료일 경우에도 비교적 안정적인 결과를 제공하는 반면, 확률 가중치가 정수로만 제한되는 단점을 갖고 있다. 그리고 대수 우도함수를 이용하여 매개변수를 추정하게 되는 최우도법은 가장 효율적인 매개변수 추정치를 얻을 수 있는 것으로 알려져 있으나, 비선형 연립방정식으로 표현되는 해를 구하기 위해서는 Newton-Raphson 방법을 사용하는 등 절차가 복잡하며, 때로는 수렴이 되지 않아 해룰 구하지 못하는 경우가 발생되게 된다. 이에 반해, 최근의 Genetic Algorithm, Ant Colony Optimization 및 Simulated Annealing과 같은 Meta-Heuristic Algorithm들은 복잡합 공학적 최적화 문제 있어서 효율적인 대안으로 주목받고 있으며, Hassanzadeh et al.(2011)에 의해 수문학적 빈도해석을 위한 매개변수 추정에 있어서도 그 적용성이 검증된바 있다. 본 연구의 목적은 연 최대강수 자료의 빈도해석에 적용되는 확률분포모형들의 매개변수 추정을 위해 Meta-Heuristic Algorithm을 적용하고자 함에 있다. 따라서 본 연구에서는 매개변수 추정을 위한 방법으로 Genetic Algorithm 및 Harmony Search를 적용하였고, 그 결과를 최우도법에 의한 결과와 비교하였다. GEV 분포를 이용하여 Simulation Test를 수행한 결과 Genetic Algorithm을 이용하여 추정된 매개변수들은 최우도법에 의한 결과들과 비교적 유사한 분포를 나타내었으나 과도한 계산시간이 요구되는 것으로 나타났다. 하지만 Harmony Search를 이용하여 추정된 매개변수들은 최우도법에 의한 결과들과 유사한 분포를 나타내었을 뿐만 아니라 계산시간 또한 매우 짧은 것으로 나타났다. 또한 국내 74개소의 강우관측소 자료와 Gamma, Log-normal, GEV 및 Gumbel 분포를 이용한 실증연구에 있어서도 Harmony Search를 이용한 매개변수 추정은 효율적인 매개 변수 추정치를 제공하는 것으로 나타났다.

• The Journal of the Korea Contents Association
• /
• v.9 no.6
• /
• pp.1-8
• /
• 2009

There are a lot of influences, such as location of camera, luminosity, brightness, and direction of light, which affect the performance of 2-dimensional image recognition. This paper suggests an adaptive method for face-image recognition in noisy environments using evolvable filtering and feature extraction which uses one sample image from camera. This suggested method consists of two main parts. One is the environmental-adjustment module which determines optimum sets of filters, filter parameters, and dimensions of features by using "steady state genetic algorithm". The other another part is for face recognition module which performs recognition of face-image using the previous results. In the processing, we used Gabor wavelet for extracting features in the images and k-Nearest Neighbor method for the classification. For testing of the adaptive face recognition method, we tested the adaptive method in the brightness noise, in the impulse noise and in the composite noise and verified that the adaptive method protects face recognition-rate's rapidly decrease which can be occurred generally in the noisy environments.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.455-455
• /
• 2011

개발로 인한 환경변화는 관련 분석모형을 통해 직접적으로 예측하기 하는 것이 가장 바람직하지만 데이터 취득의 어려움, 분석 방법론의 부재 등의 이유로 정량적 평가가 어려운 현실이다. 그렇기 때문에 수자원사업을 계획시 대부분 환경적인 영향을 매우 정성적인 형태로 평가하거나 수질과 같은 대표적인 항목에 대해서만 예측하는 수준이다. 기존의 연구 또한, 유역 또는 행정구역의 현재의 현 상황을 평가하기 위한 것이 주이며, 수자원사업과 관련성이 적은 항목도 일부 포함되어 있기 때문에 수자원사업의 특수성을 반영하기에 한계가 있다. 현 상황의 이러한 문제점을 인식하여 본 연구는 오늘날 대표적 의사결정 기법이라 할 수 있는 계층화분석과정(AHP)과 다속성효용이론(MAUT)을 활용하여 향후 수자원사업과 관련된 다기준의 사결정 과정에서의 환경성 평가방안을 제시하였다. 환경성 평가기준은 수질, 경관, 생태계 이렇게 세 가지 항목으로 구성하였고, 각 평가기준에 대한 수준을 직접적으로 대변 가능한 정량화 방안을 제시하였다. 그리고 앞서 정량화된 값을 표준화하기 위하여 MAUT 기법으로부터 평가기준별 효용함수를 도출하였다. 한편, 사업을 시행함에 따라 예상되는 환경성변화는 사업전 환경성과 사업 후 환경성을 비교하도록 하였고, 이때 해당사업의 특수성을 반영하고자 별도의 설문과정을 통해 평가기준별 가중치를 결정하였다. 본 연구는 환경성 검토시 생태학적, 물리적 분석에 기반을 둔 정량적 예측의 어려움을 보완하기 위해 정성적 예측을 추가적으로 제시하였고, 사업의 특수성과 평가항목이 갖는 일반성을 명확히 구분하여 의사결정 과정에서 주관적인 요소를 최소화하였다. 또한, 평가항목별 사업전후의 환경성을 비교, 검토함으로써 실제 사업추진 과정에서 개발로 인한 부정적 영향의 사전예방에 도움을 줄 수 있을 것으로 판단된다.

• Journal of Navigation and Port Research
• /
• v.43 no.1
• /
• pp.16-22
• /
• 2019

This study proposed estimation methods for weight and center of gravity of small fishing vessels. Weights loaded on small fishing vessels were divided into fixed weights such as crew, fishing gear, and variable weights such as fuel, fresh water, provision, bait, and fish. Based on statistical analyses with weight data of several small fishing vessels, weight, longitudinal center of gravity (LCG), vertical center of gravity (KG) of each item were represented as linear functions of vessel gross tonnage. In addition, weighting factors of variable weights were added on estimation formulas in accordance with vessel loading conditions, e.g. full load departure condition. Estimation methods were verified using actual small fishing vessel data.

• Journal of Digital Convergence
• /
• v.17 no.5
• /
• pp.217-223
• /
• 2019

A study in predicting defects of spot welding in real time in automotive field is essential for cost reduction and high quality production. Welding quality is determined by shear strength and the size of the nugget, and results depend on different independent variables. In order to develop the real-time prediction system, multiple regression analyses were conducted and the two dependent variables were obtained with sufficient statistical results with three independent variables, however, the quality prediction by the regression formula could not ensure accuracy. In this study, a multi-layer neural network circuit was constructed. The neural network by 10 dynamic resistance variables was constructed with three hidden layers to obtain execution functions and weighting matrix. In this case, the neural network was established with three independent variables based on regression analysis, as there could be difficulties in real-time control due to too many input variables. As a result, all test data were divided into poor, partial, and modalities. Therefore, a real-time welding quality determination system by three independent variables obtained by multiple regression analysis was completed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.714-720
• /
• 2021

The ITL (information theoretic learning) based on the kernel density estimation method that has successfully been applied to machine learning and signal processing applications has a drawback of severe sensitiveness in choosing proper kernel sizes. For the maximization of correntropy criterion (MCC) as one of the ITL-type criteria, several methods of adapting the remaining kernel size ( ) after removing the term have been studied. In this paper, it is shown that the main cause of sensitivity in choosing the kernel size derives from the term and that the adaptive adjustment of in the remaining terms leads to approach the absolute value of error, which prevents the weight adjustment from continuing. Thus, it is proposed that choosing an appropriate constant as the kernel size for the remaining terms is more effective. In addition, the experiment results when compared to the conventional algorithm show that the proposed method enhances learning performance by about 2dB of steady state MSE with the same convergence rate. In an experiment for channel models, the proposed method enhances performance by 4 dB so that the proposed method is more suitable for more complex or inferior conditions.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1022-1027
• /
• 2020

In this paper, SVM-based speaker classification is experimented with GMM-supervector. To create a speaker cluster, conventional speaker change detection is performed with the KL distance using the SNR-based weighting function. SVM-based speaker classification consists of two steps. In the first step, SVM-based classification between UBM and speaker models is performed, speaker information is indexed in each cluster, and then grouped by speaker. In the second step, the SVM-based classification between UBM and speaker models is performed by inputting the speaker cluster group. Linear and RBF are applied as kernel functions for SVM-based classification. As a result, in the first step, the case of applying the linear kernel showed better performance than RBF with 148 speaker clusters, MDR 0, FAR 47.3, and ER 50.7. The second step experiment result also showed the best performance with 109 speaker clusters, MDR 1.3, FAR 28.4, and ER 32.1 when the linear kernel was applied.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.882-897
• /
• 2020

Recently, the use of immersive media contents representing Panorama and 360° video is increasing. Since the viewing angle is limited to generate the content through a general camera, image stitching is mainly used to combine images taken with multiple cameras into one image having a wide field of view. However, if the parallax between the cameras is large, parallax distortion may occur in the stitched image, which disturbs the user's content immersion, thus an image stitching overcoming parallax distortion is required. The existing Seam Optimization based image stitching method to overcome parallax distortion uses energy function or object segment information to reflect the location information of objects, but the initial seam generation location, background information, performance of the object detector, and placement of objects may limit application. Therefore, in this paper, we propose an image stitching method that can overcome the limitations of the existing method by adding a weight value set differently according to the type of object to the energy value using object detection based on deep learning.