• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1727-1731
• /
• 2004

We propose an ART2 neural network-based fault diagnosis method to diagnose of sensor in the gas monitoring system. In the proposed method, using thermal modulation of operating temperature of sensor, the signal patterns are extracted from the voltage of load resistance. Also, fault classifier by ART2 NN (adaptive resonance theory 2 neural network) with uneven vigilance parameters is used for fault isolation. The performances of the proposed fault diagnosis method are shown by simulation results using real data obtained from the gas monitoring system.

• Progress in Medical Physics
• /
• v.33 no.4
• /
• pp.142-149
• /
• 2022

Purpose: This study seeks to compare the dosimetric parameters of the bulk electron density (ED) approach and synthetic computed tomography (CT) image in terms of position variation of the air cavity in magnetic resonance-guided radiotherapy (MRgRT) for patients with pancreatic cancer. Methods: This study included nine patients that previously received MRgRT and their simulation CT and magnetic resonance (MR) images were collected. Air cavities were manually delineated on simulation CT and MR images in the treatment planning system for each patient. The synthetic CT images were generated using the deep learning model trained in a prior study. Two more plans with identical beam parameters were recalculated with ED maps that were either manually overridden by the cavities or derived from the synthetic CT. Dose calculation accuracy was explored in terms of dose-volume histogram parameters and gamma analysis. Results: The D_95% averages were 48.80 Gy, 48.50 Gy, and 48.23 Gy for the original, manually assigned, and synthetic CT-based dose distributions, respectively. The greatest deviation was observed for one patient, whose D_95% to synthetic CT was 1.84 Gy higher than the original plan. Conclusions: The variation of the air cavity position in the gastrointestinal area affects the treatment dose calculation. Synthetic CT-based ED modification would be a significant option for shortening the time-consuming process and improving MRgRT treatment accuracy.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.18 no.1
• /
• pp.42-49
• /
• 2009

This study proposes a web-based remote monitoring system for evaluating degradation of machine tools using ART2(Adaptive Resonance Theory 2) neural network. A number of studies on the monitoring of machine tools using neural networks have been reported. However, when normal condition is changed due to factors such as maintenance, tool change etc., or a new failure signal is generated, such algorithms need to be entirely retrained in order to accommodate the new signals. To cope with such problems, this study develops a remote monitoring system using ART2 in which new signals when required are simply added to the classes previously trained. This system can monitor degradation as well as failure of machine tools. To show the effectiveness of the proposed approach, the system is experimentally applied to monitoring a simulator similar to the main spindle of a machine tool, and the results show that the proposed system can be extended to monitoring of real industrial machine tools and equipment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.830-837
• /
• 2004

Neural networks-based fault diagnosis algorithm to detect and isolate faults in the nonlinear systems is proposed. In the proposed method, the fault is detected when the errors between the system output and the multilayer neural network-based nominal model output cross a Predetermined threshold. Once a fault in the system is detected, the system outputs are transferred to the fault classifier by nultilayer/ART2 NN (adaptive resonance theory 2 neural network) for fault isolation. From the computer simulation results, it is verified that the proposed fault diagonal method can be performed successfully to detect and isolate faults in a nonlinear system.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.706-712
• /
• 2009

In this paper, a fault diagnosis method is proposed to detect and classifies faults that occur in press process line. An oil level automatic monitoring method is also presented to detect oil level. The FFT(fast fourier transform) frequency analysis and ART2 NN(adaptive resonance theory 2 neural network) with uneven vigilance parameters are used to achieve fault diagnosis in proposing method, and GUI(graphical user interface) program for fault diagnosis and oil level automatic monitoring using LabVIEW is produced and fault diagnosis was done. The experiment results demonstrate the effectiveness of the proposed fault diagnosis method of induction motors and oil level automatic monitor system.

• Phonetics and Speech Sciences
• /
• v.1 no.2
• /
• pp.3-8
• /
• 2009

In an effort to improve speech recognition, we investigated the effect of the number of clusters. In usual LBG clustering, the number of codebook clusters is doubled on each bifurcation and hence cannot be chosen arbitrarily in a natural way. To have the number of clusters at our control, we combined adaptive resonance theory (ART2) with LBG and perform the clustering in two stages. The codebook thus formed was used in subsequent processing of fuzzy vector quantization (FVQ) and HMM for speech recognition tests. Compared to conventional LBG, our method was shown to reduce the best recognition error rate by 0${\sim$}0.9% depending on the vocabulary size. The result also showed that between 400 and 800 would be the optimal number of clusters in the limit of small and large vocabulary speech recognitions of isolated words, respectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10b
• /
• pp.1570-1575
• /
• 1991

The feasibility of using an adaptive resonance network (ART2) with unsupervised learning capability for too] wear detection in turning operations is investigated. Specifically, acoustic emission (AE) and cutting force signals were measured during machining, the multichannel AR coefficients of the two signals were calculated and then presented to the network to make a decision on tool wear. If the presented features are significantly different from previously learned patterns associated with a fresh tool, the network will recognize the difference and form a new category m worn tool. The experimental results show that tool wear can be effectively detected with or without minimum prior training using the self-organization property of the ART2 network.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.90-93
• /
• 2006

본 논문에서는 자율신경망 중 ART2(Adaptive Resonance Theory 2)를 이용하여 지문의 매칭알고리즘에 적용하였다. 지문의 영상을 센서로부터 입력 받아, 전 처리와 후처리 과정을 거친 후 각각의 지문에 대한 특징값을 구하고, 지문 영상을 분류 및 매칭 할 수 있도록 하였다. 다음으로 제시한 알고리즘을 바탕으로 PC(Personal Computer) 없이 독립적으로 사용 할 수 있는 실시간 임베디드 지문 인식 시스템을 구현 하였다. 실시간 임베디드 지문 인식 시스템 설계에 있어 크기와 기능면을 고려해 메인 모듈의 프로세서로 최근 신호 처리에 많이 사용되고 있는 DSP(Digital Signal Processor)를 사용 하였으며, 지문을 입력 받기 위한 센서로는 반도체 지문 센서를 사용하였다. 메인 모듈과 센서를 가지고 간단한 디스플레이 및 통신 테스트를 위해 PIC Micro-Processor를 사용해 컨트롤 보드를 제작하여 간단한 인식 테스트를 하였다. 제작한 보드를 가지고 다양한 어플리케이션이 가능하나, 본 논문에서는 하드웨어나 소프트웨어 개발에 사용 가능한 RDK(Reference Design Kit)를 최종으로 구현하였다.

• Proceedings of the KOSOMBE Conference
• /
• v.1997 no.11
• /
• pp.228-231
• /
• 1997

In this paper, we designed an ART(Adaptive Resonance Theory) network based on LPC(Linear Prediction Coefficient) for classification of PVB (Premature Ventricular Beat: PVC, LBBB, RBBB). The procedure of proposed system consists of the error calculation, feature generation and processing of the ART network. The error is calculated after processing by linear prediction algorithm and the features of ART network or classification are obtained from the binary ata determined by threshold method. In conclusion, ART network has good performance in classification of PVB.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.781-784
• /
• 2004

본 시스템은 어항속에 있는 물고기 움직임을 추적하기 위해 두 대의 카메라로부터 동시에 독립된 영상을 획득하고 획득된 영상을 처리하여 좌표를 얻어내고 3차원 좌표로 생성해내는 시스템이다. 제안하는 방법은 크게 두 대의 카메라로부터 동시에 영상을 획득하는 방법과 획득된 영상에 대한 처리 및 물체 위치 검출, 그리고 3차원 좌표 생성으로 구성된다. Frame grabber를 사용하여 두 개의 카메라로부터 동시에 영상을 획득하며, 3개의 연속된 프레임에 대한 차영상과 ART2(Adaptive Resonance Theory)를 이용하여 각각의 영상에서의 물고기 위치를 검출한다. 검출된 각각의 좌표를 병합하여 3차원 좌표를 생성하며, 추적 결과는 OpenGL을 이용하여 3차원으로 재생한다.