• Journal of information and communication convergence engineering
• /
• 제7권4호
• /
• pp.419-430
• /
• 2009

Reading comprehension (RC) systems are to understand a given text and return answers in response to questions about the text. Many previous studies extract sentences that are the most similar to questions as answers. However, texts for RC tests are generally short and facts about an event or entity are often expressed in multiple sentences. The answers for some questions might be indirectly presented in the sentences having few overlapping words with the questions. This paper proposes a conceptual graph matching method towards RC tests to extract answer strings. The method first represents the text and questions as conceptual graphs, and then extracts subgraphs for every candidate answer concept from the text graph. All candidate answer concepts will be scored and ranked according to the matching similarity between their sub-graphs and question graph. The top one will be returned as answer seed to form a concise answer string. Since the sub-graphs for candidate answer concepts are not restricted to only covering a single sentence, our approach improved the performance of answer extraction on the Remedia test data.

• 인간행동과 음악연구
• /
• 제2권2호
• /
• pp.69-79
• /
• 2005

This study was purported to examine the definition of tone deafness, various factors for the cause based on literature review of research findings, and to examine therapeutic application of music for treatment of tone deafness. With research, it was found that there can be three different kinds of tone deafness; amusia, agnosia, and asonia. Literature review showed that tone deafness has been frequently dealt in many research in order to verify the causal factors, such as gender, age, and environments. With time, the research trend on tone deafness has shifted towards neurological approach closely examining brain activity, presenting the statement that the brain's capacity to perceive modest pitch changes may be congenitally impaired. Also physiological factors contribute to tone deafness called diplacusis, which is a phenomenon wherein a given tone is heard as different pitches by the two ears, resulting in conflicting bilateral perception of pitch. Music can be used for treatment of various factors causing tone deafness. The most efficient intervention was singing program. Pitch-matching training can be effective training using operant conditioning procedure. Successive approximation or reinforcement of correct response alone was more efficient procedure in helping uncertain singers to sing on pitch. Also progressive breathing exercises helped the training the pitch-matching where one had to coordinate hearing and voice.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제22권2호
• /
• pp.91-100
• /
• 2018

In this paper, we propose a new method for template matching, patch ratio, to inpaint unknown pixels. Before this paper, many inpainting methods used sum of squared differences(SSD) or sum of absolute differences(SAD) to calculate distance between patches and it was very useful for closest patches for the template that we want to fill in. However, those methods don't consider about geometric similarity and that causes unnatural inpainting results for human visuality. Patch ratio can cover the geometric problem and moreover computational cost is less than using SSD or SAD. It is guaranteed about finding the most similar patches by Cauchy-Schwarz inequality. For ignoring unnecessary process, we compare only selected candidates by priority calculations. Exeperimental results show that the proposed algorithm is more efficent than Criminisi's one.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1988년도 전기.전자공학 학술대회 논문집
• /
• pp.529-532
• /
• 1988

This paper proposes a graph matching algorithm based on simulated annealing, which assures the globally optimal solution for circuit partitioning for the placement in the rectilinear region occurring as a result of the pre-placement of some macro cells, or onto the nonplanar surface in some military or space applications. The circuit graph ($G_{C}$) denoting the circuit topology is formed by a hierarchical bottom-up clustering of cells, while another graph called region graph ($G_{R}$) represents the geometry of a planar rectilinear region or a nonplanar surface for circuit placement. Finding the optimal many-to-one vertex mapping function from $G_{C}$ to $G_{R}$, such that the total mismatch cost between two graphs is minimal, is a combinatorial optimization problem which was solved in this work for various examples using simulated annealing.

• 전기학회논문지
• /
• 제56권1호
• /
• pp.1-7
• /
• 2007

Generally, an electric railroad feeding system has many problems due to the different characteristics in contrast with a load of general three-phase AC electric power system. One of them is harmonics problem caused by the switching device existing in the feeding system, and moreover, the time-varying dynamic loads of rail way is inherently another cause to increase this harmonics problem. In Korea power systems, the electric railroad feeding system is directly supplied from the substation of KEPCO. Therefore, if voltages fluctuation or unbalanced voltages are created by the voltage and current distortion or voltage drop during operation, it affects directly the source of supply. The trainloads of electric railway system have non-periodic but iterative harmonic characteristics as operating condition, because the electric characteristic of the electric railroad feeding system is changed by physical conditions of the each trainload. According to the traditional study, the estimation of harmonics has been performed by deterministic way using the steady state data at the specific time. This method is easy to analyze harmonics, but it has limits in some cases which needs an assessment of dynamic load and reliability. Therefore, this paper proposes the probabilistic estimation method, moments matching method(MW) in order to overcome the drawback of deterministic method. In this paper, distributions for each harmonics are convolved to obtain the moments and cumulants of TDD(Total Demand Distortion), and this can be generalized for any number of trains. For the case study, the electric railway system of LAT(Intra Airport Transit) in Incheon International Airport is modeled using PSCAD/EMTDC dynamic simulator. The raw data of harmonics for the moments matching method is acquired from simulation of the LAT model.

• 전자공학회논문지S
• /
• 제35S권12호
• /
• pp.55-62
• /
• 1998

두개의 영상을 정합 하는 것은 많은 컴퓨터 시각장치의 응용과정 중 기본적인 과정이다. 본 논문에서는 선형특징을 사용한 정합기법으로서 회전각도와 크기비율에 불변한 영상정합 기법을 제안한다. 영상은 edge 검출, 세선화, 선형화 과정에 의해 선형 세그먼트의 집합으로 묘사된다. 세그먼트 사이의 각도차이와 새로운 거리척도에 의한 크기비율을 사용해 Hough 공간에서 최대로 일치하는 변환 파라메터를 추정한다. 추정된 파라메터는 1단계 relaxation과 Hough 기법으로 이루어진 고속 선형특징 정합과정에 의해 검증된다. 제안한 기법은 변환 파라메터에 대한 사전정보를 필요로 하지 않으며 추출된 선형 세그먼트 크기의 변화에 민감하지 않은 특성과 기존의 relaxation 기법에 비해 빠른 처리속도를 가진다.

• 한국위성정보통신학회논문지
• /
• 제9권1호
• /
• pp.57-64
• /
• 2014

최근 물리적 디바이스의 도움 없이 사람이 시스템과 인터랙션 할 수 있는 인간 친화적인 인간-기계 인터페이스가 많이 연구되고 있다. 이중 대표적인 것이 본 논문의 주제인 비전기반 제스처인식이다. 본 논문에서 우리는 설정된 가상세계의 객체와의 인터랙션을 위한 손 제스처들을 정의하고 이들을 인식할 수 있는 효과적인 방법론을 제안한다. 먼저, 웹캠으로 촬영된 저해상도 영상에서 사용자의 양손을 검출 및 추적하고, 손 영역을 분할하여 손 실루엣을 추출한다. 우리는 손 검출을 위해, RGB 공간에서 명암에 따라 두개의 타원형 모델을 이용하여 피부색을 모델링하였으며, 블랍매칭(blob matching) 방법을 이용하여 손 추적을 수행하였다. 우리는 플러드필(floodfill) 알고리즘을 이용해 얻은 손 실루엣의 행/열 모드 검출 및 분석을 통해 Thumb-Up, Palm, Cross 등 세 개의 손모양을 인식하였다. 그리고 인식된 손 모양과 손 움직임의 콘텍스트를 분석해서 다섯 가지 제스처를 인식할 수 있었다. 제안하는 제스처인식 방법론은 정확한 손 검출을 위해 카메라 앞에 주요 사용자가 한 명 등장한다는 가정을 하고 있으며 많은 실시간 데모를 통해 효율성 및 정확성이 입증되었다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제8권5호
• /
• pp.187-192
• /
• 2019

일차원 셀룰러 오토마타(CA)에서 과반수 문제(majority problem)와 동기화 문제(synchronization problem)는 국소 정보(local information)를 이용하여 전역 문제(global problem)를 풀어야 하는 계산적으로 어려운 문제이다. 본 논문에서는 일차원 CA에서 과반수 문제와 동기화 문제를 푸는 CA의 규칙을 찾는 방법을 제안한다. CA의 상태전이 함수(state transition function)를 일반적으로 사용하는 규칙표(rule table)가 아닌 조건부 매칭 규칙(CMR)으로 나타내고 진화 알고리즘을 적용하였다. 각 문제에서 다수의 규칙들을 찾아내어 제안한 방법을 효과적으로 사용할 수 있음을 보였다. 또한 이전 연구 결과와 비교하여 과반수 문제와 동기화 문제에서 CMR을 사용하는 방식의 효용성을 보였으며, 다른 일차원 CA문제에도 CMR을 활용할 수 있는 가능성을 보였다.

• 산업경영시스템학회지
• /
• 제39권4호
• /
• pp.60-66
• /
• 2016

Compound logistics is a service aimed to enhance logistics efficiency by supporting that shippers and consigners jointly use logistics facilities. Many of these services have taken place both domestically and internationally, but the joint logistics services for e-commerce have not been spread yet, since the number of the parcels that the consigners transact business is usually small. As one of meaningful ways to improve utilization of compound logistics, we propose a brokerage service for shipper and consigners based on the hybrid recommendation system using very well-known classification and clustering methods. The existing recommendation system has drawn a relatively low satisfaction as it brought about one-to-one matches between consignors and logistics vendors in that such matching constrains choice range of the users to one-to-one matching each other. However, the implemented hybrid recommendation system based brokerage agent service system can provide multiple choice options to mutual users with descending ranks, which is a result of the recommendation considering transaction preferences of the users. In addition, we applied feature selection methods in order to avoid inducing a meaningless large size recommendation model and reduce a simple model. Finally, we implemented the hybrid recommendation system based brokerage agent service system that shippers and consigners can join, which is the system having capability previously described functions such as feature selection and recommendation. As a result, it turns out that the proposed hybrid recommendation based brokerage service system showed the enhanced efficiency with respect to logistics management, compared to the existing one by reporting two round simulation results.

• 한국HCI학회논문지
• /
• 제10권2호
• /
• pp.45-55
• /
• 2015

최근 증강현실 분야에서 특징점 인식 및 추적 기술은 비마커 기반의 증강 현실 서비스 구현에 중요한 역할을 담당하고 있다. 특징점 인식 및 추적 기술은 오래 전부터 컴퓨터 비전 등 여러 분야의 많은 연구자들에 의해 심도 있게 연구되어 왔으며, 특히 최근 급성장하고 있는 모바일 관련 시스템에 적용하기 위해 모바일 임베디드 환경에 접목 가능한 특징점 기반의 다양한 인식 및 추적 기술들이 소개되고 있다. 따라서 본 논문에서는 널리 활용되고 있는 특징점 기반의 매칭 및 추적의 다양한 핵심 요소 기술 (특징점 추출, 특징점 기술, 특징점 매칭 및 추적)에 대한 최신 동향을 분석하고, 본 한국과학기술연구원 연구팀이 수행한 모바일 증강현실 서비스 관련 사례 연구인 관광 지도 인식 및 추적 연구를 소개하고자 한다.