• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2585-2587
• /
• 2003

한전에서는 전력설비의 효율적인 관리를 위하여 일찌기 배전분야에 GIS를 도입하여 신배전정보시스템(NDIS : New Distribution Information System)을 구축, 시범운영을 마치고 전국에 단계적으로 확대운영하고 있다. GIS를 업무에 활용하기 위해서는 설비도면의 입력이 선행되어야 하나 이를 수작업에 의존할 경우 많은 비용과 시간이 소요될 뿐만 아니라 입력자의 숙련정도에 따라 자료의 정확도가 달라지게 되므로, 이러한 문제점들을 근본적으로 해결하고자 설비의 위치, 심볼, 계통연결, 속성자료 등을 컴퓨터로 자동인식 입력시켜 수작업을 최소화하는 기법 및 적용연구가 필요하며, 특히 국가기본도를 Base Map으로 사용함에 따른 상대오차 보정문제도 해결되어야 한다. 본 개발은 변전소에서 전력수용가까지의 전력공급설비를 나타내는 배전설비도면에서 도면내 주요 설비인 전주와 전선을 인식하는 방법 즉, 반투명 필름에 손으로 그려진 배전설비도면의 스캐닝 영상을 인식기법을 적용하여 설비내용, 설치위치, 전선종류별 설치상태 등 지리정보시스템에서 사용될 정보를 Digital 형태의 Data로 자동생성하고 국가기본도와의 상대오차보정까지 처리하는 것을 주요내용으로 하고 있다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.517-525
• /
• 2006

In this paper, the algorithm which improve the efficiency of surveillance in spite of the change of light is proposed and confirmed by virtue of the experiments. One of the problems for the implementation of visual surveillance system is the image processing technique to overcome with the variations of illumination lights. Some conventional systems are generally not considered the error due to the change of lights because the system use at indoor. In practical, the factors of bad image can be classified to the ghosts due to the reflection of lights and shadows in a scene. Especially weak images and noises at night are decreased the performance of visual surveillance system. In the paper, the filter which improve the images with some change of illumination lights is designed and the gabor filter is used for recognition and tracking of the moving objects. In the results, the system showed that the recognition and tracking were obtained $92\sim100%$ of recognition rate at daytime, but $80\sim90%$ of nighttime.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.691-694
• /
• 2011

차량의 위치를 인식하기 위한 시스템으로는 위성 위치 확인 시스템(GPS, Global Positioning System)과 관성 항법 시스템(INS, Inertial Navigation System)이 있다. INS는 차량의 최초 위치를 입력해야한다는 점과 시간이 지남에 따라 오차가 누적된다는 점 때문에 GPS와 INS가 상호보완적인요소로 통합하여 사용되고 있다. 하지만 GPS로부터 얻는 위치 정보는 정확성의 문제가 존재한다. 본 논문에서는 이를 해결하기 위해 톨게이트와 전광판을 활용하여 초기의 위치 값을 얻고, 누적오차를 방지하기 위해 재 초기화하는 방안을 제안한다. 고속도로상의 톨게이트와 전광판에는 모두 DSRC(Dedicate Short Range Communication) 시스템을 통해 위치를 전송할 수 있다. 따라서 INS의 최초 위치 입력이 필요한 문제와 누적오차 문제를 해결할 수 있다. 제안 방식을 통해 INS의 장점을 살리면서도 좀 더 정확한 위치를 인식 할 수 있어 차량 간 통신(V2V, Vehicle-to-Vehicle)기반의 이웃 차량 위치인지 시스템에 대한 연구가 더 활발해질 것으로 기대된다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.7
• /
• pp.1439-1446
• /
• 2011

This paper describes the algorithm that lowers the dimension, maintains the object recognition and significantly reduces the eigenspace configuration time by combining the higher correlation feature information and Principle Component Analysis. Since the suggested method doesn't require a lot of computation than the method using existing geometric information or stereo image, the fact that it is very suitable for building the real-time system has been proved through the experiment. In addition, since the existing point to point method which is a simple distance calculation has many errors, in this paper to improve recognition rate the recognition error could be reduced by using several successive input images as a unit of recognition with K-Nearest Neighbor which is the improved Class to Class method.

• Journal of Korea Multimedia Society
• /
• v.13 no.3
• /
• pp.400-409
• /
• 2010

This paper presents a position error recognition system when the wafer is mounted in cleaning equipment among the wafer manufacturing processes. The proposed system is to enhance the performance in cost and reliability by preventing the wafer cleaning system from damaging by alerting it when it is put in correct position. The key algorithms are the calibration method between image acquired from camera and physical wafer, a infrared lighting and the design of the filter, and the extraction of wafer boundary and the position error recognition resulting from generation of circle based on least square method. The system is to install in-line process using high reliable and high accurate position recognition. The experimental results show that the performance is good in detecting errors within tolerance.

• Journal of Animal Environmental Science
• /
• v.17 no.2
• /
• pp.93-100
• /
• 2011

This study was conducted to verify the applicability of robot milking system through acquisition and analysis of model teat's position information using scanning range finder (SRF). Model teats, same size and shape as real teats, were designed to analyze the properties according to the material, distance error and angle error of the sensor. In addition, 2-dimensional distance information of each teats was obtained at same time with 4 teat models and the result were as follows. 1. In the case of the fingers on the experiment for selection of materials for teat model, the distance error was from 4.3 mm to 1.3 mm, average was 2.8 mm as a minimum record. In the case of rubber material, average distance error was 4.3 mm. So, this material was considered to be a most suitable model. 2. The distance error was maximum at 100 mm distance. The more distance increased, the less error increased up to 300 mm. Then the error increased after 300 mm and decreased again. 3. The maximum angle error of 10.1 mm was measured at $170^{\circ}$, in case of $70^{\circ}$ the error was 0.2 mm as a minimum value. There was no specific tendency to error of angle. 4. In the 2-dimensional location error for 4 teat models, distance error was 3.8 mm as minimum and 7.2 mm as maximum. The angle error was $1.2^{\circ}$ as maximum. All of errors were included within the accuracy of sensor, the robot milking system was considered to be applicable to measure the distance of teats due to the measuring velocity of SRF and the hole size of teat-cup.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.613-615
• /
• 2011

본 논문에서는 켑스트럼 계수를 이용하여 음성인식을 하는 알고리즘을 제안한다. 본 논문에서 제안하는 방법은 사람이 발성한 음성을 두 영역의 켑스트럼 계수로 분리한 후에, 신경회로망을 사용하여 음성인식을 하는 방법이다. 본 논문에서 제안하는 신경회로망은 오차가 거의 없어지는 일정 기간 동안 네트워크를 학습시킨 후에 신경회로망의 학습 데이터와는 다른 새로운 음성이 신경회로망에 입력된 경우에 대하여 각 음성 구간에서 분류가 가능한 모음검출을 위한 음성인식 시스템을 제안한다.

• Journal of Advanced Marine Engineering and Technology
• /
• v.37 no.2
• /
• pp.213-219
• /
• 2013

Recently, with the rapid advancement in information and communication technology, indoor location-based services(LBSs) that require precise position tracking have been actively studied with outdoor-LBS using GPS. However, in case of a ship which consists of steel structure, it is difficult to measure a precise localization due to significant ranging error by the diffraction and refraction of radio waves. In order to reduce location measurement errors that occur in these indoor environments, this paper presents distance compensation algorithms that are suitable for a narrow passage such as ship corridors without any additional sensors by using UWB(Ultra-wide-band), which is robust to multi-path and has an error in the range of a few centimeters in free space. These improvement methods are that Pythagorean theory and adjacent node technique are used to solve the distance error due to the node deployment and distance variation analysis technique is applied to reduce the ranging errors which are significantly fluctuated in the corner section. The experimental results show that the number of nodes and the distance error are reduced to 66% and 57.41%, respectively, compared with conventional CSS(Chirp spread spectrum) method.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.83-86
• /
• 2012

최근 임베디드 시스템이 이용되는 이동체나 자동화 공장에서 각 디바이스들간의 분배된 작업이 정확한 시간에 맞추어 작동 할 것을 요구하고 있어 시간 동기화는 중요하게 인식되고 있다. 각각의 시스템은 자기의 시계를 가지고 처리되기 때문에 이들간의 시간 정보의 오차를 적게 하여야 한다. 이에 본 논문에서는 SNTP 프로토콜을 이용해 타임서버로부터 시간의 단위를 정하여 단위 간격마다 시간을 여러 번 확인하여 오차를 조정하고, 실험을 통해 평균오차와 최대오차를 구하여 시간 동기화 시점을 확인하였다. 이를 통해 시간 동기화 시점을 미리 조정하여 다음 장치들 간의 통신 시점에서 동기화를 할 경우 오차를 줄이는 방법을 제안한다. 본 논문에서 제안한 방법을 Rabbit Board 6700 으로 구현하고 검증하여 이를 시간적 정밀함이 요구되는 응용분야에 적용할 수 있게 한다.

• Journal of the Korean Data and Information Science Society
• /
• v.9 no.2
• /
• pp.345-355
• /
• 1998

The back propagation(BP) algorithm allows multilayer feedforward neural networks to learn input-output mappings from training samples. It iteratively adjusts the network parameters(weights) to minimize the sum of squared approximation errors using a gradient descent technique. However, the mapping acquired through the BP algorithm may be corrupt when errorneous training data are employed. In this paper two types of robust backpropagation algorithms are discussed both from a theoretical point of view and in the case studies of nonlinear regression function estimation and handwritten Korean character recognition. For future research we suggest Bayesian learning approach to neural networks and compare it with two robust backpropagation algorithms.