• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권8호
• /
• pp.742-751
• /
• 2005

본 논문은 단어 분류에 기반한 새로운 텍스트 영상 워터마킹 알고리즘을 제안한다. 간단한 특징을 이용하여 단어를 K개로 분류한다. 이웃한 몇 개의 단어들을 조합하여 세그먼트를 구성하고, 세그먼트에 속한 단어들의 부류에 의해 세그먼트 또한 분류된다. 각 세그먼트에 동일한 양의 신호가 삽입된다. 신호 삽입은 세그먼트 부류가 갖는 단어 간 공백의 통계값을 조작함으로써 이루어진다. 몇 가지 기준에 따라 기존 단어 이동 알고리즘과의 주관적인 비교가 제시된다.

• 한국멀티미디어학회논문지
• /
• 제15권12호
• /
• pp.1430-1441
• /
• 2012

In this paper, we present an algorithm that quickly and effectively estimates orthogonal vanishing points in equirectangular images of urban environment. Our algorithm is based on the RANSAC (RANdom SAmple Consensus) algorithm and on the characteristics of the line segment in the spherical panorama image of the $360^{\circ}$ longitude and $180^{\circ}$ latitude field of view. These characteristics can be used to reduce the geometric ambiguity in the line segment classification as well as to improve the robustness of vanishing point estimation. The proposed algorithm is validated experimentally on a wide set of images. The results show that our algorithm provides excellent levels of accuracy for the vanishing point estimation as well as line segment classification.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권8호
• /
• pp.335-343
• /
• 2002

A segment matching algorithm in stereo vision via the fusion of multiple features on long edge segments is proposed. One problem of the previous segment matching algorithm is the similarity among the segments caused from its short length. In the proposed algorithm, edges are composed of longer segments which are obtained by breaking the edges only at the locations with distinguished changes of the shape. Such long segments can contain extra features such as curvature ratio and length of segments which could not be included in shorter ones. Use of such additional features enhances the matching accuracy significantly To fuse multiple features for matching, weighting value determination algorithm which is computed according to the degree of the contribution of each factor is proposed. The stereo matching simulations with the proposed algorithm are done about various images and their results are included.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 학술회의 논문집 정보 및 제어부문 B
• /
• pp.1022-1030
• /
• 2003

In this paper we propose a efficient offset curve construction algorithm for $C^0$-continuous Open and Closed 2D sequence curve with line segment in the plane. One of the most difficult problems of offset construction is the loop problem caused by the interference of offset curve segments. Prior work[1-10] eliminates the formation of local self-intersection loop before constructing a intermediate(or raw) offset curve, whereas the global self-intersection loop are detected and removed explicitly(such as a sweep algorithm[13]) after constructing a intermediate offset curve. we propose an algorithm which removes global as well as local intersection loop without making a intermediate offset curve by forward tracing of tangential circle. Offset of both open and closed poly-line segment sequence curve in the plane constructs using the proposed approach.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.176-180
• /
• 2004

The ST-segment that the beginning part of T wave is the important diagnostic parameter to finding myocardial ischemia. Abnormal ST appears in two types. One is the level change, and the other is the pattern change. In this paper, we describe the monitoring of abnormal ST using PC based system. Hardware of this system consists of transmitter, receiver and PC. The function of transmitter is measuring ECG in three channels which are selected manually and transmitting the data to receiver by digital radio way. Connection with receiver and PC is by RS232C, and the data received on the PC is analyzed automatically by ECG analysis algorithm and saved to file. In the algorithm part for detecting abnormal ST, ST-segments are approximated by a polynomial. This method can detect all of the deviation and pattern change of ST-segment regardless the change in the heart rate or sampling rate. To gain algorithm reliability, the method rejects distorted polynomial approximation by calculation the difference between the approximated ST-segment and original ST-segment. In pre-signal processing, the wavelet transformation separates high frequency bands including QRS complex from the original ECG. Consequently, the process improves the performance of detecting each feature points.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 추계학술대회 논문집 학회본부 A
• /
• pp.200-202
• /
• 1999

This paper describes the optimization problem of real power rescheduling and present an algorithm based linear programming for studying the load-shedding and generation reallocation problem when a portion of the transmission system is disabled and at power flow solution cannot be obtained for the overload of some lines. And in case initial is infeasible, solution could not be converge. So this paper gives an algorithm being lie infeasible quantities within limit. The paper describes a LP-based algorithm to obtain the solution in power dispatch related to overload situations in power system and it is easily extened under various objective. The optimization procedures is based in linear programming with bounded variables and use the multi-segment curve method for a objective function and the validity of the algorithm is verified with two examples : 10-bus system and 57-bus system.

• 대한임베디드공학회논문지
• /
• 제17권5호
• /
• pp.297-307
• /
• 2022

The simple procedural segment selection algorithm commonly used in Dynamic Adaptive Streaming over HTTP (DASH) reveals severe weakness to provide high-quality streaming services in the integrated mobile networks of various wired and wireless links. A major issue could be how to properly cope with dynamically changing underlying network conditions. The key to meet it should be to make the segment selection algorithm much more adaptive to fluctuation of network traffics. This paper presents a system architecture that replaces the existing procedural segment selection algorithm with a deep reinforcement learning algorithm based on the Asynchronous Advantage Actor-Critic (A3C). The distributed A3C-based deep learning server is designed and implemented to allow multiple clients in different network conditions to stream videos simultaneously, collect learning data quickly, and learn asynchronously, resulting in greatly improved learning speed as the number of video clients increases. The performance analysis shows that the proposed algorithm outperforms both the conventional DASH algorithm and the Deep Q-Network algorithm in terms of the user's quality of experience and the speed of deep learning.

• 한국멀티미디어학회논문지
• /
• 제11권3호
• /
• pp.292-300
• /
• 2008

본 논문에서는 점진적인 화소 확장을 이용하여 영상 내에 존재하는 선분을 찾아내는 선분 추출 알고리즘을 제안한다. 본 논문에서 제안하는 선분 추출 방법에서는 기존의 선분 추출 방법인 허프 변환 기반 방법이나 선분의 그룹화 기반 방법과는 다른 접근법을 사용하였다. 영상이 입력되면 케니 테두리를 구하고, 테두리 화소 중 임의의 점을 선택하여 선분을 근사화 시킨 기본 직선을 만든 후, 선택된 점에서 임의의 반경 내에 있는 테두리 화소들을 선택한다. 직선과의 거리 오차와 기울기 각의 오차를 이용하여 선택된 화소에 가중치를 부여한다. 가중치 합 비교에 의해 선택된 화소들이 떨어져 있는지를 판별한 후, 가중치를 적용한 최소자승법에 의해 선 맞춤을 하여 선분을 구하게 된다. 제안된 알고리즘은 기존에 제안된 방법들과 결과를 비교하였으며, 계산 속도가 빠르면서 실제 존재하는 선분 추출이 가능하다는 것을 실험 결과를 통해 제시한다.

• 한국멀티미디어학회논문지
• /
• 제22권2호
• /
• pp.225-230
• /
• 2019

There are several methods which efficiencies of database are uprise. One of the well-known methods is that segments of database satisfying a query was rapidly accessed and processed. So if it is possible to search completely parallel multiple database segment types which satisfy a query, the response time of the query will be reduced. The matter of obtaining CPS(Completely Parallel Searchable) distribution without redundancy can be viewed as graph theoretic problem, and the operation of ring sum on the graph is used for CPS. In this paper, the parallel algorithm is proposed.

• 한국정보처리학회논문지
• /
• 제6권8호
• /
• pp.2213-2221
• /
• 1999

직선 선분은 선분의 중점에 대하여 대칭 성질을 갖고 있다. 우리는 이 대칭성에 착안하여 직선 생성을 위한 픽셀 선택시 한 번에 두 개의 픽셀을 동시에 선택할 수 있는 수정 브레제남 알고리즘을 제안하였다. 본 알고리즘은 한번에 두 개의 픽셀 위치를 결정하며, 픽셀 별로 각각의 결정 파라미터(decision parameter)를 사용하는 것이 아니라 오직 한 개의 결정 파라미터만을 사용한다. 이러한 선택 방법을 뒷받침하기 위한 이론적 증명을 아울러 제시하였다. 본 논문에서 제안한 알고리즘은 픽셀 선택을 위한 루프 수행 회수가 줄어들기 때문에 직선 생성에 걸리는 시간을 약 5% 내외로 단축할 수 있음을 실험을 통하여 알 수 있었다. 또한, 생성된 직선의 모양은 기존 브레제남 알고리즘의 그것과 일치함을 확인할 수 있었다.