• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.2
• /
• pp.101-108
• /
• 2001

본 논문에서는 배전선로에서 발생하는 여러 고장유형을 판별하기 위해서 적응형 퍼지추론 시스템을 적용하는 새로운 기법을 제시하였다. 배전선로의 고장과 고장유사현상 데이터를 추출하기 위해서 EMTP를 이용하여 RL부하, 아크로부하, 컨버터부하가 있는 배전계통을 구성하고 여러 형태의 고장과 고장유사현상에 대해 시뮬레이션을 하였다. 이를 통해 얻은 전류 파형으로부터 기본파성분, 영상분전류, 짝수 고조파성분의 합, 홍수 고조파성분의 합, 그리고 비정규 고조파성분의 합의 5개의 입력변수를 추출하고 학습을 통해서 각 입력변수의 소속함수의 소속도를 자동으로 결정하였다. 이 적응형 퍼지추론 시스템을 이용한 기법을 평가하기 위해서 학습시와 다른 고장상황을 모의하여 얻은 데이터와 실증시험 데이터를 이용하였다. 결과적으로 제안한 기법은 배전선로에서 발생하는 고장유형을 빠르고 정확하게 판별할 수 있었다.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.237-248
• /
• 1992

본 논문에서는 통계적인 특징 추출 방법을 사용하여 샘물체로 필기된 한글을 고속으로 인식하는 방법을 소개한다. 대부분이 직선 성분으로 이루어진 한글의 특성을 이응하기 위하여 입력 영상으로부터 수평, 수직, 사선, 역사선의 방향 성분을 추출하며, 검은 화소의 밀도에 따라 동적으로 그물을 결정함으로써 획 간의 접촉 변형에 무관한 특징 벡터를 추출한다. 이와 같은 통계적 특징 추출 방법은 크기 정규화나 세선화 과정이 필요없으며, 또한 샘물체라는 필기 형태의 제약에 의해 정합 대상 부류의 수가 현저히 줄어들기 때문에 인식에 소요되는 시간을 상당히 줄일 수 있음은 물론, 인식률을 향상시켰다. 제안된 방법의 타당성을 검증하기 위하여 샘물체로 필기된 KS 완성형 한글 2,350자에 대해 실험한 결과, 평균 90% 이상의 인식률을 보이며, IBM PC 486(33MHz)상에서 문자당 평균 0.17초의 인식 속도를 보임으로써, 실용적인 고속 OCR 시스템의 개발 가능성을 확인할 수 있었다.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.7
• /
• pp.901-914
• /
• 2002

In this paper, We propose an efficient automatic caption detection and location method, caption recognition using FE-MCBP(Feature Extraction based Multichained BackPropagation) neural network for content based retrieval of video. Frames are selected at fixed time interval from video and key frames are selected by gray scale histogram method. for each key frames, segmentation is performed and caption lines are detected using line scan method. lastly each characters are separated. This research improves speed and efficiency by color segmentation using local maximum analysis method before line scanning. Caption detection is a first stage of multimedia database organization and detected captions are used as input of text recognition system. Recognized captions can be searched by content based retrieval method.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1135-1147
• /
• 2021

In order to geometrically correct high-resolution satellite imagery, the sensor modeling process that restores the geometric relationship between the satellite sensor and the ground surface at the image acquisition time is required. In general, high-resolution satellites provide RPC (Rational Polynomial Coefficient) information, but the vendor-provided RPC includes geometric distortion caused by the position and orientation of the satellite sensor. GCP (Ground Control Point) is generally used to correct the RPC errors. The representative method of acquiring GCP is field survey to obtain accurate ground coordinates. However, it is difficult to find the GCP in the satellite image due to the quality of the image, land cover change, relief displacement, etc. By using image maps acquired from various sensors as reference data, it is possible to automate the collection of GCP through the image matching algorithm. In this study, the RPC of KOMPSAT-3A satellite image was corrected through the extracted matching point using the UAV (Unmanned Aerial Vehichle) imagery. We propose a pre-porocessing method for the extraction of matching points between the UAV imagery and KOMPSAT-3A satellite image. To this end, the characteristics of matching points extracted by independently applying the SURF (Speeded-Up Robust Features) and the phase correlation, which are representative feature-based matching method and area-based matching method, respectively, were compared. The RPC adjustment parameters were calculated using the matching points extracted through each algorithm. In order to verify the performance and usability of the proposed method, it was compared with the GCP-based RPC correction result. The GCP-based method showed an improvement of correction accuracy by 2.14 pixels for the sample and 5.43 pixelsfor the line compared to the vendor-provided RPC. In the proposed method using SURF and phase correlation methods, the accuracy of sample was improved by 0.83 pixels and 1.49 pixels, and that of line wasimproved by 4.81 pixels and 5.19 pixels, respectively, compared to the vendor-provided RPC. Through the experimental results, the proposed method using the UAV imagery presented the possibility as an alternative to the GCP-based method for the RPC correction.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.2
• /
• pp.69-80
• /
• 1999

Recently, social requirements of personal identification techniques are rapidly expanding in a number of new application ares. Especially fingerprint recognition is the most important technology. Fingerprint recognition technologies are well established, proven, cost and legally accepted. Therefore, it has more spot lighted among the any other biometrics technologies. In this paper we propose a new on-line fingerprint recognition algorithm for non-inked type live scanner to fit their increasing of security level under the computing environment. Fingerprint recognition system consists of two distinct structural blocks: feature extraction and feature matching. The main topic in this paper focuses on the feature matching using the fingerprint minutiae (ridge ending and bifurcation). Minutiae matching is composed in the alignment stage and matching stage. Success of optimizing the alignment stage is the key of real-time (on-line) fingerprint recognition. Proposed alignment algorithm using clique shows the strength in the search space optimization and partially incomplete image. We make our own database to get the generality. Using the traditional statistical discriminant analysis, 0.05% false acceptance rate (FAR) at 8.83% false rejection rate (FRR) in 1.55 second average matching speed on a Pentium system have been achieved. This makes it possible to construct high performance fingerprint recognition system.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.3211-3218
• /
• 1997

This paper addresses an eye location algorithm which is essential process of human face tracking system for natural video-conferencing. In current video-conferencing systems, user's facial movements are restricted by fixed camera, therefore it is inconvenient to users. We Propose an eye location algorithm for automatic face tracking. Because, locations of other facial features guessed from locations of eye and scale of face in the image can be calculated using inter-ocular distance. Most previous feature extraction methods for face recognition system are approached under assumption that approximative face region or location of each facial feature is known. The proposed algorithm in this paper uses no prior information on the given image. It is not sensitive to backgrounds and lighting conditions. The proposed algorithm uses the valley representation as major information to locate eyes. The experiments have been performed for 213 frames of 17 people and show very encouraging results.

• Journal of Korea Multimedia Society
• /
• v.5 no.3
• /
• pp.255-262
• /
• 2002

A boundary sequence can be a good representation of arbitrary shapes, because it can represent them simply and precisely. However, boundary sequences have not been used as a representation of arbitrary shapes, because the pixel-based shape-features such as area, centroid, orientation, projection and so forth, could not be computed directly from them. In this paper, we show that the shape-features can be easily computed from the boundary sequences by introducing the cross-sections that are defined as vertical (or horizontal) line segments in a shape. A cross-section generation method is proposed, which generates cross-sections of the shape efficiently by tracing the boundary sequence of the shape once. Furthermore, a boundary sequence extraction method is also proposed, which generates a boundary sequence for each shape in a binary image automatically The proposed methods work well even if a shape has holes. Eventually, we show that a boundary sequence can be used effectively for representing arbitrary shapes.

• Proceedings of the KSRS Conference
• /
• 1998.09a
• /
• pp.186-191
• /
• 1998

The first step in change detection in any SAR monitoring, including SAR interferometry, is the co-registration of the images. CCPs (Ground Control Points) for co-registration are usually detected manually, but for qualitative analyses of enormous volumes of data, some automation of the process will become necessary. An automated determination of common CCPs for the same path/row data is especially desirable. We selected the intersections of linear features as the candidates of common GCPs Very bright point targets, which are commonly used as GCPs, have the drawback of appearing and disappearing depending on the conditions of the observation. But in the case of linear features, some detailed elements may appear differently in some case, but the overall line-likeness will remain. In this study, we selected 18 common GCPs for a single-look JERS-1 SAR image of Omaezaki area in central Japan. Although the GCPs in the first image had to be selected either interactively or semi-automatically, the same GCPs in all other images were successively detected automatically using a tiny sub-image around each GCP and a dilated mask of each linear feature in the first image as the reference data.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.3
• /
• pp.97-105
• /
• 2017

This study is the realization study recognizing the Korean gothic printing letter. This study defined the new grapheme by using the connection ingredient and had the graphemes recognized by means of the feature dots of the isolated dot, end dot, 2-line gathering dots, more than 3 lines gathering dots, and classified the characters by means of the arrangement information of the graphemes and the layers that the graphemes form within the characters, and made the character database for the recognition by using them. The layers and the arrangement information of the graphemes consisting in the characters were presumed by using the weight center position information of the graphemes extracted from the characters to recognize and the information of the graphemes obtained by vertically exploring from the weight center of each grapheme, and it recognized the characters by judging and comparing the character groups of the database by means of the information which was secured this way. 350 characters were used for the character recognition test and about 97% recognition result was obtained by recognizing 338 characters.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1054-1061
• /
• 2003

Q-Learning, a representative algorithm of reinforcement learning, experiences repeatedly until estimation values about all state-action pairs of state space converge and achieve optimal policies. When the state space is high dimensional or continuous, complex reinforcement learning tasks involve very large state space and suffer from storing all individual state values in a single table. We introduce Q-Map that is new function approximation method to get classified policies. As an agent learns on-line, Q-Map groups states of similar situations and adapts to new experiences repeatedly. State-action pairs necessary for fine control are treated in the form of rule. As a result of experiment in maze environment and mountain car problem, we can achieve classified knowledge and extract easily rules from Q-Map