• Communications for Statistical Applications and Methods
• /
• 제8권2호
• /
• pp.543-556
• /
• 2001

This study suggests integrated neural network modes for he stock price index forecasting using change-point detection. The basic concept of this proposed model is to obtain significant intervals occurred by change points, identify them as change-point groups, and reflect them in stock price index forecasting. The model is composed of three phases. The first phase is to detect successive structural changes in stock price index dataset. The second phase is to forecast change-point group with various data mining classifiers. The final phase is to forecast the stock price index with backpropagation neural networks. The proposed model is applied to the stock price index forecasting. This study then examines the predictability of integrated neural network models and compares the performance of data mining classifiers.

• Structural Engineering and Mechanics
• /
• 제77권2호
• /
• pp.151-165
• /
• 2021

This paper aims to present a novelty damage detection method to identify damage locations by the simultaneous use of both the energy and displacement damage indices. Using this novelty method, the damaged location and even the damaged floor are accurately detected. As a first method, a combination of the instantaneous frequency energy index (EDI) and the structural acceleration responses are used. To evaluate the first method and also present a rapid assessment method, the Displacement Damage Index (DDI), which consists of the error reliability (β) and Normal Probability Density Function (NPDF) indices, are introduced. The innovation of this method is the simultaneous use of displacement-acceleration responses during one process, which is more effective in the rapid evaluation of damage patterns with velocity vectors. In order to evaluate the effectiveness of the proposed method, various damage scenarios of the ASCE benchmark problem, and the effects of measurement noise were studied numerically. Extensive analyses show that the rapid proposed method is capable of accurately detecting the location of sparse damages through the building. Finally, the proposed method was validated by experimental studies of a six-story steel building structure with single and multiple damage cases.

• 대한공간정보학회지
• /
• 제21권2호
• /
• pp.63-69
• /
• 2013

변화탐지 기법은 위성영상의 활용 및 국토 모니터링에 있어서 필수적인 알고리즘이다. 그러나, 변화탐지 기법을 고해상도 위성영상에 적용할 경우, 다시기 영상 간의 기하학적 차이 등에 의하여 변화탐지 정확도가 저하될 수 있다. 본 연구에서는 효과적인 위성영상의 변화탐지를 위하여 기존의 융합 영상 평가지수를 활용하고자 한다. 또한, 기존의 다시기 위성영상을 활용한 일반적인 변화탐지 기법과 교차융합영상을 이용한 변화탐지 결과를 비교하여, 다시기 고해상도 위성영상에 적합한 변화탐지 기법을 제안하고자 한다. 이를 위해, 융합영상 평가 지수인 ERGAS, UIQI, SAM를 무감독 변화탐지 기법에 적용하고 기존의 CVA를 이용한 변화탐지 기법의 결과와 비교하였다. 또한, 영상융합 기법에 따른 고해상도 위성영상 변화탐지 정확도를 평가하여 고해상도 위성영상의 무감독 변화탐지에서 발생할 수 있는 기하학적 오차를 최소화할 수 있는 방법을 분석하였다. 실험결과, 교차융합영상과 ERGAS 지수를 활용한 변화탐지 기법이 기존 기법과 비교하여 상대적으로 높은 변화지역 탐지 가능성을 가지는 것을 확인할 수 있었다.

• 제어로봇시스템학회논문지
• /
• 제22권4호
• /
• pp.276-280
• /
• 2016

Due to the increasing interest in safety and consistent product quality over a past few decades, demand for effective quality monitoring and safe operation in the modern industry has propelled research into statistical based fault detection and diagnosis methods. This paper describes the application of Hotelling $T^2$ index based Principal Component Analysis (PCA) method for fault detection and diagnosis in industrial processes. Multivariate statistical process control techniques are now widely used for performance monitoring and fault detection. Conventional methods such as PCA are suitable only for steady state processes. These conventional projection methods causes false alarms or missing data for the systems with transient values of processes. These issues significantly compromise the reliability of the monitoring systems. In this paper, a reliable method is used to overcome false alarms occur due to varying process conditions and missing data problems in transient states. This monitoring method is implemented and validated experimentally along with matlab. Experimental results proved the credibility of this fault detection method for both the steady state and transient operations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.662-675
• /
• 2018

Compression history detection plays an important role in digital multimedia forensics. Most existing works, however, mainly focus on digital image and video. Additionally, the existed audio compression detection algorithms aim to detect the trace of double compression. In real forgery scenario, multiple compression is more likely to happen. In this paper, we proposed a detection algorithm to reveal the compression history for MP3 audio. The statistics of the scale factor and Huffman table index which are the parameters of MP3 codec have been extracted as the detecting features. The experimental results have shown that the proposed method can effectively identify whether the testing audio has been previously treated with single/double/triple compression.

• 한국자동차공학회논문집
• /
• 제7권5호
• /
• pp.1-8
• /
• 1999

This paper describes the method for detection of the misfired cylinder by analysis of the variation of pressure occurred in exhaust manifold on an MPI gasoline engine. Misfired cylinder(s) cause a loss of power, an increase of fuel consumption and exhaust emission and vibration is caused by unsteady torque. Therefore early detection and correction of misfired cylinder(s) play a very important role in the proper performance and the exhaust emission. The method is a comparison of integration pressure index during the period of a blowdown in the displacement period. Experimental results showed that the method, using the variation of pressure in the exhaust manifold is proven to be effective in the detection of single cylinder or multiple cylinders misfire on the gasoline engine regardless of the engine revolutions. In addition, this method, using the variation of pressure in the exhaust manifold is a very easy and accurate method compared with other methods.

• Communications for Statistical Applications and Methods
• /
• 제10권1호
• /
• pp.153-166
• /
• 2003

This study presents the time-divided forecasting model to integrate evolutionary optimization algorithm and change point detection based on artificial neural networks (ANN) for the prediction of (Korea) stock price index. The genetic algorithm(GA) is introduced as an evolutionary optimization method in this study. The basic concept of the proposed model is to obtain intervals divided by change points, to identify them as optimal or near-optimal change point groups, and to use them in the forecasting of the stock price index. The proposed model consists of three phases. The first phase detects successive change points. The second phase detects the change-point groups with the GA. Finally, the third phase forecasts the output with ANN using the GA. This study examines the predictability of the proposed model for the prediction of stock price index.

• Journal of Communications and Networks
• /
• 제18권3호
• /
• pp.439-445
• /
• 2016

In this paper, we consider subcarrier-index modulation (SIM) for precoded orthogonal frequency division multiplexing (OFDM) with a few activated subcarriers per user and its generalization to multi-carrier multiple access systems. The resulting multiple access is called sparse index multiple access (SIMA). SIMA can be considered as a combination of multi-carrier code division multiple access (MC-CDMA) and SIM. Thus, SIMA is able to exploit a path diversity gain by (random) spreading over multiple carriers as MC-CDMA. To detect multiple users' signals, a low-complexity detection method is proposed by exploiting the notion of compressive sensing (CS). The derived low-complexity detection method is based on the orthogonal matching pursuit (OMP) algorithm, which is one of greedy algorithms used to estimate sparse signals in CS. From simulation results, we can observe that SIMA can perform better than MC-CDMA when the ratio of the number of users to the number of multi-carrier is low.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.144-144
• /
• 2022

Identifying waterbody from remote sensing images, namely water detection, helps understand continuous redistribution of terrestrial water storage and accompanying hydrological processes. It also allows us to estimate available surface water resources and help effective water management. For this problem, NDWI (Normalized Difference Water Index) and MNDWI (Modified Normalized Difference Water Index) are widely used. Although remote sensing indexes can highlight remote sensing image in the water, the noise and the spatial information of the remote sensing image are difficult to be considered, so the accuracy is difficult to be compared with the visual interpretation (the most accurate method, but it requires a lot of labor, which makes it difficult to apply). In this study, we attempt to improve existing NDWI and MNDWI to better water detection. We establish waterbody database of South Korea first and then used it for assessing waterbody indices.

• 정보과학회 논문지
• /
• 제45권3호
• /
• pp.294-302
• /
• 2018

정보통신 기술의 발전에 따른 새로운 서비스 산업의 출현으로 개인 정보 침해, 산업 기밀 유출 등 사이버 공간의 위험이 다양화 되어, 그에 따른 보안 문제가 중요한 이슈로 떠오르게 되었다. 본 연구에서는 기업 내 개인 정보 오남용 및 내부 정보 유출에 따른, 대용량 사용자 로그 데이터를 기반으로 기존의 시그니처(Signature) 보안 대응 방식에 비해, 실시간 및 대용량 데이터 분석기술에 적합한 행위 기반 이상 탐지방식을 제안하였다. 행위 기반 이상 탐지방식이 대용량 데이터를 처리하는 기술을 필요로 함에 따라, 역방향 인덱스(Inverted Index) 기반의 실시간 검색 엔진인 엘라스틱서치(Elasticsearch)를 사용하였다. 또한 데이터 분석을 위해 통계 기반의 빈도 분석과 전 처리 과정을 수행하였으며, 밀도 기반의 군집화 방법인 DBSCAN 알고리즘을 적용하여 이상 데이터를 분류하는 방법과 시각화를 통해 분석을 간편하게 하기위한 한 사례를 보였다. 이는 기존의 이상 탐지 시스템과 달리 임계값을 별도로 설정하지 않고 이상 탐지 분석을 시도하였다는 것과 통계적인 측면에서 이상 탐지 방식을 제안하였다는 것에 의의가 있다.