• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.2
• /
• pp.78-88
• /
• 2009

In many face recognition problems, the number of available images is limited compared to the dimension of the input space which is usually equal to the number of pixels. This problem is called as the 'small sample size' problem and regularization methods are typically used to solve this problem in feature extraction methods such as LDA. By using regularization methods, the modified within class matrix becomes nonsingu1ar and LDA can be performed in its original form. However, in the process of adding a scaled version of the identity matrix to the original within scatter matrix, the scale factor should be set heuristically and the performance of the recognition system depends on highly the value of the scalar factor. By using the proposed resampling method, we can generate a set of images similar to but slightly different from the original image. With the increased number of images, the small sample size problem is alleviated and the classification performance increases. Unlike regularization method, the resampling method does not suffer from the heuristic setting of the parameter producing better performance.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.852-859
• /
• 2009

Normally, to extract the defect in TFT-LCD inspection system, the image is obtained by using line scan camera or area scan camera which is achieved by CCD or CMOS sensor. Because of the limited dynamic range of CCD or CMOS sensor as well as the effect of the illumination, these images are frequently degraded and the important features are hard to decern by a human viewer. In order to overcome this problem, the feature vectors in the image are obtained by using the average intensity difference between defect and background based on the weber's law and the standard deviation of the background region. The defect detection method uses non-linear SVM (Supports Vector Machine) method using the extracted feature vectors. The experiment results show that the proposed method yields better performance of defect classification methods over conveniently method.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.6
• /
• pp.1-8
• /
• 2009

Detecting entry and exit zones in a view covered by multiple cameras is an essential step to determine the topology of the camera setup, which is critical for achieving and sustaining the accuracy and efficiency of multi-camera surveillance system. In this paper, a graph theoretic clustering method is proposed to detect zones using data points which correspond to entry and exit events of objects in the camera view. The minimum spanning tree (MST) is constructed by associating the data points. Then a set of well-formed clusters is sought by removing inconsistent edges of the MST, based on the concepts of the cluster balance and the cluster density defined in the paper. Experimental results suggest that the proposed method is effective, even for sparsely elongated clusters which could be problematic for expectation-maximization (EM). In addition, comparing to the EM-based approaches, the number of data required to obtain stable outcome is relatively small, hence shorter learning period.

• Journal of Digital Contents Society
• /
• v.9 no.3
• /
• pp.389-398
• /
• 2008

There has been some research into a speech system for communications between those who are hearing impaired and those who hear normally, but the system has been pursued in inefficient teaching ways in which existing teachers teach each individual due to social indifference and a lack of marketability. In order to overcome such a weakness, there appeared to be a need to develop contents utilizing 3D animation and digital technology. For the investigation of a standard face and a standard spherical shape for the preparation of a character, the study collected sufficient data concerning students in the third-sixth grades in elementary schools in Seoul and Gyeonggi, Korea, and drew up standards for a face and a spherical shape of such students. This data is not merely the basic data of content development for the hearing impaired, but it can also offer a standard measurement and a standard type realistically applicable to them. As a system for understanding conversations by applying 3D character animation and educating self-expression, the character speech animation system supports effective learning for language education for hearing impaired children who need language education within their families and in special education institutions with the combination of 3D technology and motion capture.

• The Journal of Society for e-Business Studies
• /
• v.9 no.4
• /
• pp.103-116
• /
• 2004

It is needed to transfer the technology actively which has already developed to improve a up-to-date technology and foster the technological innovation. The technology transfer also can bring about a commercial success. To promote the technology transfer, it is needed to develop a new technology valuation model for a specific technology from a objective point of view, as well as to equip an institution such as the technology transfer center. The technology valuation from a objective point of view is of importance as the basic information for the price negotiation between a technology-buyer and a technology-seller. This paper takes aim at investigating a new technology valuation model and developing a technology valuation system for promoting the technology transfer. A new technology valuation system is developed as a web-enabling base. Using this users are able to estimate the value of specific technology on a real time efficiently.

• Archives of design research
• /
• v.15 no.3
• /
• pp.149-156
• /
• 2002

Through academic-industrial collaboration program, special knowledge and human resources in the university are utilized in the society and it may contribute to the development of industry and enhance the competitiveness of the corporation to the activation and acceleration of the regional economy and finally to the enhancement of competitiveness of national industry in international level. This study is based upon design process for the development of body-shower product for bath room, and expected to be used as an important educational material for establishment of order-based education system which conforms to the job site needs that may be analysed from feedback of product results and for practical learning.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.5
• /
• pp.359-368
• /
• 2013

Mashup is a web application that combines several different sources to create new services using Open APIs(Application Program Interfaces). Although the mashup has become very popular over the last few years, there are several challenging issues when combining a large number of APIs into the mashup, especially when composite APIs are manually integrated by mashup developers. This paper proposes a novel algorithm for automatic Open API composition. The proposed algorithm consists of constructing an operation connecting graph and searching composition candidates. We construct an operation connecting graph which is based on the semantic similarity between the inputs and the outputs of Open APIs. We generate directed acyclic graphs (DAGs) that can produce the output satisfying the desired goal. In order to produce the DAGs efficiently, we rapidly filter out APIs that are not useful for the composition. The algorithm is evaluated using a collection of REST and SOAP APIs extracted from ProgrammableWeb.com.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.1
• /
• pp.13-21
• /
• 2020

Surrogate models have been used for the rapid estimation of six-DOF aerodynamic coefficients in the context of the design and control of a missile. For this end, we may generate highly accurate surrogate models with a multitude of aerodynamic data obtained from wind tunnel tests (WTTs); however, this approach is time-consuming and expensive. Thus, we aim to swiftly predict aerodynamic coefficients via co-Kriging using a few WTT data along with plenty of computational fluid dynamics (CFD) data. To demonstrate the excellence of co-Kriging models based on both WTT and CFD data, we first generated two surrogate models: co-Kriging models with CFD data and Kriging models without the CFD data. Afterwards, we carried out numerical validation and examined predictive trends to compare the two different surrogate models. As a result, we found that the co-Kriging models produced more accurate aerodynamic coefficients than the Kriging models thanks to the assistance of CFD data.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.8A
• /
• pp.786-795
• /
• 2002

In this paper, we present a histogram and moment-based vidoe scencd change detection technique using hierarchical Hidden Markov Models(HMMs). The proposed method extracts histograms from a low-frequency subband and moments of edge components from high-frequency subbands of wavelet transformed images. Then each HMM is trained by using histogram difference and directional moment difference, respectively, extracted from manually labeled video. The video segmentation process consists of two steps. A histogram-based HMM is first used to segment the input video sequence into three categories: shot, cut, gradual scene changes. In the second stage, a moment-based HMM is used to further segment the gradual changes into a fade and a dissolve. The experimental results show that the proposed technique is more effective in partitioning video frames than the previous threshold-based methods.

• Journal of Internet Computing and Services
• /
• v.13 no.4
• /
• pp.1-10
• /
• 2012

In this paper, an optimal feature selection method considering sematic of features, which is preprocess of document classification is proposed. The feature selection is very important part on classification, which is composed of removing redundant features and selecting essential features. LSA (Latent Semantic Analysis) for considering meaning of the features is adopted. However, a supervised LSA which is suitable method for classification problems is used because the basic LSA is not specialized for feature selection. We also apply GA (Genetic Algorithm) to the features, which are obtained from supervised LSA to select better feature subset. Finally, we project documents onto new selected feature subset and classify them using specific classifier, SVM (Support Vector Machine). It is expected to get high performance and efficiency of classification by selecting optimal feature subset using the proposed hybrid method of supervised LSA and GA. Its efficiency is proved through experiments using internet news classification with low features.