• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.2
• /
• pp.165-165
• /
• 2012

육상, 해양, 항공 등의 응용분야에 위성항법보강시스템의 활용을 위해서는 시스템의 정확성, 무결성, 연속성, 가용성 요구 조건을 만족하도록 설계되어야 하며, 무결성 요구 조건을 만족시키기 위하여 측위 오차 및 위협 요인들을 지상국에서 감시해야한다. 특히, 전리층 변화는 지역적으로 경향 및 세기가 달라 전리층 폭풍 발생 시 지상국과 이동체에서 받은 위성항법 신호에 포함된 전리층 지연 오차의 편차가 심하여 위성항법 사용자의 무결성, 즉 안정성이 위협을 받는 상황이 발생할 수 있으므로, 해당지역의 전리층 변화에 대한 사전 정보를 통해 지역별로 적합한 위협 모델을 구성하여 전리층 활동 감시가 필요하다. 전리층 기울기는 전리층 지연값 분포의 불균일 여부를 정량화한 값으로, 전리층 폭풍 발생시 기울기가 급증하여 전리층 폭풍 감지를 위한 지표로 활용될 수 있다. 이 연구에서는 육상 교통 위성항법보강시스템의 무결성 감시에 전리층 변화 기능을 적용하기 위한 기본 연구로 IGS에서 제공하는 전리층 보정정보를 이용하여 한반도 상공에 대한 전리층 기울기 분포 및 변화 경향을 파악하고, 이러한 분석 결과를 전리층 기울기에 대한 보정정보 오차범위 설정이나 전리층 폭풍 발생 판단에 필요한 임계값 설정 등에 적용하고자 한다.

• Journal of Korea Water Resources Association
• /
• v.41 no.5
• /
• pp.527-544
• /
• 2008

Generalized Pareto distribution (GPD) is frequently applied in hydrologic extreme value analysis. The main objective of statistics of extremes is the prediction of rare events, and the primary problem has been the estimation of the threshold and the exceedances which were difficult without an accurate method of calculation. In this paper, to obtain the threshold or the exceedances, four methods were considered. For this comparison a GPD model was used to estimate parameters and quantiles for the seven durations (1, 2, 3, 6, 12, 18 and 24 hours) and the ten return periods (2, 3, 5, 10, 20, 30, 50, 70, 80 and 100 years). The parameters and quantiles of the three-parameter generalized Pareto distribution were estimated with three methods (MOM, ML and PWM). To estimate the degree of fit, three methods (K-S, CVM and A-D test) were performed and the relative root mean squared error (RRMSE) was calculated for a Monte Carlo generated sample. Then the performance of these methods were compared with the objective of identifying the best method from their number.

• The Journal of the Korea Contents Association
• /
• v.16 no.5
• /
• pp.104-112
• /
• 2016

The Adaboost chooses a good set of features in rounds. On each round, it chooses the optimal feature and its threshold value by minimizing the weighted error of classification. The involved process of classification performs a hard decision. In this paper, we expand the process of classification to a soft fuzzy decision. We believe this expansion could allow some flexibility to the Adaboost algorithm as well as a good performance especially when the size of a training data set is not large enough. The typical Adaboost algorithm assigns a same weight to each training datum on the first round of a training process. We propose a new algorithm to assign different initial weights based on some statistical properties of involved features. In experimental results, we assess that the proposed method shows higher performance than the traditional one.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.166-172
• /
• 1998

PD 제어기는 산업현장에 가장 널리 운영되는 제어알고리즘이지만 지금까지 보편적으로 사용되고 있는 PID 파라메터 동조방법인 Ziegler-Nichols 동조법으로는 화학공정 같은 비선형 특성이 크거나, 시정수가 큰 플래트에서는 좋은 성능을 얻을 수 없다. 본 논문에서는 릴레이 동조 실험을 통하여 임계 이득과 발진주기를 구하고, 이 값들로부터 Z-N 동조법을 적용하여 초기 동조값을 구한다. 이 값에 의해서 얻어진 공정 출력의 오차와 오차변화율을 입력으로 한 퍼지 동조기를 통하여 PID 제어기의 비례이득과 적분시간을 구하는 동조 방법을 제시한다.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.10
• /
• pp.72-82
• /
• 1999

The error diffusion algorithm is excellent for reproducing continuous gray-scale images to bianry images and also has good edge characteristics. However, it is well known that artifacts with objectionable patterns can occur in the halftoned images. On the other hand, a halftone algorithm using blue noise mask has been proposed. where the halftoning is achieved by a pixelwise comparison of gray-scale image with an array, the blue noise mask. It doesn't have pattern artifacts, but the halftoned image looks unclear because the quantization errors are not feedbacked compared to the error diffusion. In this paper, edge enhanced error diffusion which dithers the threshold with the blue noise mask is proposed. We show that the proposed algorithm can produce unstructured and edge enhanced halftone images. This algorithm is analyzed by the concept of an equivalent input image. The performace of the proposed algorithm is compared with that of the conventional halftoning by measuring the radially averaged power spectrum and edge correlation.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.2
• /
• pp.24-30
• /
• 2003

In this Paper VTQM(Variable Threshold quantization Method) is proposed, which enhances PSNR by composing the advantages of round-off and truncation quantization methods. From vast experiments, PSNR's of JPEG images with the method are compared with that of standard JPEG methods. At the boundary which affects the compression ratio, VTQM quantizes with a new threshold to reduce the length of compressed data; otherwise, VTQM employs the round-off quantization method to preserve the original image information. As a result, VTQM reduces bit rate while preserving the image quality. The experimental results show that VTQM enhances PSNR by 0.2∼0.3dB, 0.4∼0.5dB with the same bit rate than those of the traditional truncation and round-off quantization methods, respectively.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.1
• /
• pp.82-87
• /
• 2001

본 논문에서는 마우스 대신에 음성으로 명령을 입력하여 퍼지 추론을 통해 위도우 화면상의 커서를 이동시키는 인터페이스를 구현하였다. 입력된 음성이 대체로 짧은 언어이기에 이를 인식하기 위하여 고립단어 인식에 강한 DTW방식을 사용하였다. DTW방식의 단점중인 하나가 음성길이가 비슷한 명령을 입력하였을 때 표준패턴 중 오차 값이 가장 작은 패턴으로 인식하는 것이다. 예를 들면 \"아주 많이 이동해\"하는 음성이 입력되었을 때 비슷한 음성길이를 가진 \"아주 많이 오른쪽\"으로 인식하는 경우가 있다. 이런 오류를 해결하고자 각 패턴의 DTW오차 거리 값과 표준 패턴의 음성길이를 기준으로 임계값을 퍼지 추론하여 명령으로서의 수락 여부를 결정하였다. 판단이 애매한 부분은 사용자에게 질의를 하여 응답에 따라 수락 여부를 결정하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.10 no.1
• /
• pp.20-29
• /
• 1991

A fundamental assumption in conventional linear predictive coding (LPC) analysis procedure is that the input to an all-pole vocal tract filter is white process. In the case of periodic inputs, however, a pitch bias error is introduced into the conventional LP coefficient. Multi-pulse (MP) LP analysis can reduce this bias, provided that an estimate of the excitation is available. Since the prediction error of conventional LP analysis can be modeled as the sum of an MP excitation sequence and a random noise sequence, we can view extracting MP sequences from the prediction error as a classical detection and estimation problem. In this paper, we propose an algorithm in which the locations and amplitudes of the MP sequences are first obtained by applying a likelihood ratio test (LRT) to the prediction error, and LP coefficients free of pitch bias are then obtained from the MP sequences. To verify the performance enhancement, we iterate the above procedure with adaptive threshold at each step.

• Journal of Aerospace System Engineering
• /
• v.12 no.6
• /
• pp.32-40
• /
• 2018

An impact localization for a carbon fiber reinforced plastic (CFRP) composite plate was performed using the multiplexed fiber bragg grating (FBG) sensors and the error-outlier based impact localization algorithm. We found that the optimal impact localization with the maximum error of 31.82 mm and the averaged error of 6.31 mm are obtained when the error threshold (ET) and constant level (CL) are 0.3 nm and 110, respectively. Moreover, the detailed process of impact localization under certain optimal parameters and the relevant results were thoroughly investigated. We conclude that the multiplexed FBG sensors and the error-outlier based impact localization algorithm are suitable for an impact localization on composite structures, and expect that they can be utilized for various structural health monitoring (SHM) in the future.