• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.6
• /
• pp.994-1003
• /
• 1994

In this paper, a scheme of noise cancellation using spectral subreaction method with single input in an autombile noise environment is proposed. In order to remove the changing automonile noise components form the noisy speech signal, the noise of various states is analyzed and its characteristics are presented. For the decision of speech/nonspeech and the estimation of noise spectrum, a classification function is proposed on the basis of noise analysis. This function presents the precise decision of speech/nonspeech and the optimal estimation of noise spectrum with less computation. As the result of the estimation of noise spectrum by the proposed classification function, the clean speech signal is extracted from the noisy speech signal with high signal-to-ratio.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.5
• /
• pp.110-116
• /
• 2008

This paper presents a new postprocessing method to eliminate the false minutiae, that caused by the skelectonization of fingerprint image, and an image compression method using Isometric Self Organizing Map(ISOSOM). Since the SOM has simple structure, fast encoding time, and relatively good classification characteristics, many image processing areas adopt this such as image compression and pattern classification, etc. But, the SOM shows limited performances in pattern classification because of it's single layer structure. To maximize the performance of the pattern classification with small code book, we a lied the Isometric SOM with the isometry of the fractal theory. The proposed Isometric SOM postprocessing and compression algorithm of fingerprint image showed good performances in the elimination of false minutiae and the image compression simultaneously.

• The Journal of the Korea Contents Association
• /
• v.21 no.8
• /
• pp.356-366
• /
• 2021

The high-tech industry, a leader in the national economy, has a larger investment cost compared to general buildings, a shorter construction period, and requires continuous investment. Therefore, accurate construction cost prediction and quick decision-making are important factors for efficient cost and process management. Overseas, the construction information classification system has been standardized since 1980 and has been continuously developed, improving construction productivity by systematically collecting and utilizing project life cycle information. At domestic construction sites, attempts have been made to standardize the classification system of construction information, but it is difficult to achieve continuous standardization and systematization due to the absence of a standardization body and differences in cost and process management methods for each construction company. Particular, in the case of the high-tech industry, the standardization and systematization level of the construction information classification system for high-tech facility construction is very low due to problems such as large scale, numerous types of work, complex construction and security. Therefore, the purpose of this study is to construct a construction information classification system suitable for high-tech facility construction through collection, classification, and analysis of related project data constructed in Korea. Based on the WBS (Work Breakdown Structure) and CBS (Cost Breakdown Structure) classified and analyzed through this study, a code system through hierarchical classification was proposed, and the cost model of buildings by linking WBS and CBS was three-dimensionalized and the utilized method was presented. Through this, an information classification system based on inter-relationships can be developed beyond the one-way tree structure, which is a general construction information classification system, and effects such as shortening of construction period and cost reduction will be maximized.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.39 no.3
• /
• pp.56-64
• /
• 2021

Pavilion is an important landscape installation in the traditional landscaping and a representative facility that organizes space. To interpret and succeed the traditional landscape space, researches on the shape and structure of the traditional pavilion are also needed. However, researches on the style and structure of the traditional pavilion are difficult to be found. Accordingly this study aimed to identify the structural characteristics of roof part that occupied the largest portion in determining the shape of pavilion. Our research findings are as follows. As a result of analyzing 15 traditional Single-room square pavilions whose structure could be identified, it was found that the main building elements that distinguished the type of roof part in the Single-room square were crossbeam, ridge pole, and baluster. Depending on the presence of roofing members, pavilion was classified into five types: crossbeam, crossbeam+ridge pole, crossbeam+baluster, ridge pole, and baluster. In addition, as a result of analyzing the load and joint that worked on crossbeam based on the shape classification of roof part, it was found that in the traditional Single-room square pavilion, crossbeam was designed to play a balancing role between pillar and load. This study is significant in that it attempted to make a close interpretation of the shape of roof part in the pavilion and the role and function of building elements in terms of size, ratio, and load.

• Tunnel and Underground Space
• /
• v.31 no.1
• /
• pp.10-24
• /
• 2021

Rock mass classification for construction of underground facilities is essential to secure their stabilities. Therefore, the reliable values for rock mass classification from the precise information on rock discontinuities are most important factors, because rock mass discontinuities can affect exclusively on the physical and mechanical properties of rock mass. The conventional classification operation for rock mass has been usually performed by hand mapping. However, there have been many issues for its precision and reliability; for instance, in large-scale survey area for regional geological survey, or rock mass classification operation by non-professional engineers. For these reasons, automated rock mass classification using LiDAR becomes popular for obtaining the quick and precise information. But there are several suggested algorithms for analyzing the rock mass discontinuities from point cloud data by LiDAR scanning, and it is known that the different algorithm gives usually different solution. Also, it is not simple to obtain the exact same value to hand mapping. In this paper, several discontinuity extract algorithms have been explained, and their processes for extracting rock mass discontinuities have been simulated for real rock bench. The application process for several algorithms is anticipated to be a good reference for future researches on extracting rock mass discontinuities from digital point cloud data by laser scanner, such as LiDAR.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.1
• /
• pp.7-13
• /
• 2023

This study investigated the criteria for evaluating the impact of artificial intelligence services. The study classified AI evaluation targets into two areas: AI service and AI technology, and identified influence, sustainability, efficiency, effectiveness, and appropriateness as potential evaluation criteria. The time aspect of AI service evaluation was divided into pre-evaluation and post-evaluation, with pre-evaluation focused on reviewing items during development and design. The AI service area was classified into public, private, and mixed forms, and the impact assessment was classified as vertical or horizontal. The application of AI services was divided into normative and regulatory aspects, and the purpose of the evaluation could be impact or process evaluation. The subject and field of the AI service could also be used for classification purposes. The results of this study can be used to support the creation of AI service impact policies and countermeasures. However, further research is needed to develop specific indicators based on the criteria identified in this study to evaluate the impact of AI services.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.04a
• /
• pp.318-321
• /
• 2000

이 연구는 정보통신기기 수출량에 관해 하향식(Top-down) 방법에 기초한 예측 모형을 제시한다. 하향식방법은 전체 수출량과 전체를 구성하는 개별 항목간에 계층적 관계를 바탕으로 순차적으로 예측을 수행하는 방법이다. 전체와 개별 항목간에 관계는 데이터의 시계열 특성과 데이터에 영향을 주는 요인들에 의해서 만들어진다. 이러한 관계를 바탕으로 하는 하향식 예측은 전체 수출량을 먼저 예측한 후 이 예측치를 바탕으로 하여 개별 항목에 대한 예측을 수행한다. 하지만 하향식 방법은 가장 아래 계층의 예측치를 산출하기 위해 필요한 것이며 최종 예측치는 가장 마지막 계층에서부터 예측 데이터를 합산해서 얻을 수 있다. 결국 하향식 예측 방법은 전체와 개별 항목 사이에 상관관계가 높고 계층화되어 있는 구조에 적합하다. 이 예측 대상이 되는 정보통신기기 수출량에 대한 적용 사례를 살펴보자. 계층 구조를 보면 정보통신기기 전체 수출량과 전체를 구성하는 개별 항목으로 정보통신기기 분류별(유선기기, 무선기기, 방송기기, 정보기기, 기타부품기기)과 국가별(미국, 일본, 중국 등 7 개국)로 나뉘어진다. 다시 이 아래 계층으로는 국가와 정보통신기기의 행렬 구조(예: 미국-유선, 일본-부품 등)에 의해 35 개로 나뉘어진다. 각 단계별 예측 방법을 보면 전체 수출량은 시계열 특성과 거시적 변수를 반영한 시계열 모형, 그 아래 계층인 국가별과 분류별 모형에는 전체 수출량 시계열 특성과 국가별과 분류별에 영향을 주는 관련 변수를 반영한 회귀모형 그리고 행렬 구조에 대한 예측은 상위 계층의 시계열 특성과 행렬구조 데이터의 계절성이 반영된 다중 회귀모형을 이용하였다.ndex, mobile user′s will first be classified by their traffic volume, and then calculate the average tariffs per minute of each group of users, and lastly weight-average those tariffs per minute. And finally, this paper shows the mobile tariff index by considering those averaged tariffs and the carriers′ market shares to reflect the contribution of individual carriers and the users′ traffic volume.완화될 수 있다. 즉, 봉지를 씌웅으로서 봉지 내의 대기 환경이 외기보다 안정적으로 유지되고 직사광선이나 농약 및 마찰로부터 과실을 보호해 주기에 동녹이 어느 정도 방지될 수 있는 것이다. 그러나 기존의 황금배봉지는 동녹의 정도를 완화시킬 뿐 완전히 방지할 수 없었으며, 봉지를 적 용한 재배조건에서의 동녹발생 기구를 정확히 이해하지 못했었기에 효과적으로 봉지의 기능 을 개선하는 것이 불가능하였다. 과설의 미려도는 과실의 맛과 함께 그 가치를 결정짓는 중요한 물성으로서 우리나라 황 금배 재배환경과 특성에 알맞은 배봉지의 제작이 선결될 때, 배 품질의 향상, 안정된 공급이 가능하게 될 것이며 아울러 농가의 수업증대와 수출 경쟁력 강화가 이루어질 수 있을 것으로 판단된다. 이러한 측면에서 황금배 재배농가가 당면한 동녹발생의 문제점을 신속한 해결 을

• The KIPS Transactions:PartB
• /
• v.8B no.3
• /
• pp.231-236
• /
• 2001

일반적인 다층 신경망에서 가중치의 갱신 알고리즘으로 사용하는 오류 역전과 방식은 가중치 갱신 결과를 고정된(fixed) 한 개의 값으로 결정한다. 이는 여러 갱신의 가능성을 오직 한 개의 값으로 고정하기 때문에 다양한 가능성들을 모두 수용하지 못하는 면이 있다. 하지만 모든 가능성을 확률적 분포로 표현하는 갱신 알고리즘을 도입하면 이런 문제는 해결된다. 이러한 알고리즘을 사용한 베이지안 신경망 모형(Bayesian Neural Networks Models)은 주어진 입력값(Input)에 대해 블랙 박스(Black-Box)와같은 신경망 구조의 각 층(Layer)을 거친 출력값(Out put)을 계산한다. 이 때 주어진 입력 데이터에 대한 결과의 예측값은 사후분포(posterior distribution)의 기댓값(mean)에 의해 계산할 수 있다. 주어진 사전분포(prior distribution)와 학습데이터에 의한 우도함수(likelihood functions)에 의해 계산한 사후확률의 함수는 매우 복잡한 구조를 가짐으로 기댓값의 적분계산에 대한 어려움이 발생한다. 따라서 수치해석적인 방법보다는 확률적 추정에 의한 근사 방법인 몬테 칼로 시뮬레이션을 이용할 수 있다. 이러한 방법으로서 Hybrid Monte Carlo 알고리즘은 좋은 결과를 제공하여준다(Neal 1996). 본 논문에서는 Hybrid Monte Carlo 알고리즘을 적용한 신경망이 기존의 CHAID, CART 그리고 QUEST와 같은 여러 가지 분류 알고리즘에 비해서 우수한 결과를 제공하는 것을 나타내고 있다.

• Journal of the Korean Society for information Management
• /
• v.21 no.1
• /
• pp.93-117
• /
• 2004

This study is to develop a hierarchic clustering model fur document classification and browsing in OPAC systems. Two automatic indexing techniques (with and without controlled terms), two term weighting methods (based on term frequency and binary weight), five similarity coefficients (Dice, Jaccard, Pearson, Cosine, and Squared Euclidean). and three hierarchic clustering algorithms (Between Average Linkage, Within Average Linkage, and Complete Linkage method) were tested on the document collection of 175 books and theses on library and information science. The best document clusters resulted from the Between Average Linkage or Complete Linkage method with Jaccard or Dice coefficient on the automatic indexing with controlled terms in binary vector. The clusters from Between Average Linkage with Jaccard has more likely decimal classification structure.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.35-40
• /
• 2009

FCM 알고리즘은 입력 벡터와 각 클러스터의 유클리드 거리를 이용하여 구해진 소속도만를 비교하여 데이터를 분류하기 때문에 클러스터링 된 공간에서의 데이터들의 분포에 따라 바람직하지 못한 클러스터링 결과를 보일 수 있다. 이러한 문제점을 개선하기 위해 대칭적 성질을 이용하는 대칭성 측도에 퍼지 이론을 적용하여 군집간의 거리에 따른 변화와 군집 중심의 위치, 그리고 군집 형태에 따라 영향을 덜 받는 개선된 FCM이 제안되었다. 본 논문에서는 효과적으로 패턴을 분류하기 위해 개선된 FCM 알고리즘을 적용한 개선된 하이브리드 네트워크를 제안한다. 제안된 하이브리드 네트워크는 개선된 FCM 알고리즘을 입력층과 중간층의 학습구조 적용하고 중간층과 출력층의 학습구조는 일반화된 델타학습법을 적용한다. 제안된 방법의 인식성능을 평가하기 위해 2차원 좌표평면 상의 데이터를 기존의 Max_Min 신경망을 이용한 FCM 기반 RBF 네트워크와 FCM 기반 RBF 네트워크, HCM 기반 네트워크와 제안된 방법 간의 학습 및 인식 성능을 비교 및 분석하였다.