• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.1133-1138
• /
• 2000

This paper describes a signal recognition method for diagnosing the rotating machinery using wavelet-aided Self-Organizing Feature Map(SOFM). The SOFM specialized from neural network is a new and effective algorithm for interpreting large and complex data sets. It converts high-dimensional data items into simple order relationships with low dimension. Additionally the Learning Vector Quantization(LVQ) is used for reducing the error from SOFM. Multi-resolution and wavelet transform are used to extract salient features from the primary vibration signals. Since it decomposes the raw timebase signal into two respective parts in the time space and frequency domain, it does not lose either information unlike Fourier transform. This paper is focused on the development of advanced signal classifier in order to automatize vibration signal pattern recognition. This method is verified by the experiment and several abnormal vibrations such as unbalance and rubbing are classified with high flexibility and reliability by the proposed methods.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1989년도 춘계학술대회
• /
• pp.19-22
• /
• 1989

This paper represents a syntactic peak detection algorithm which detects peaks in the ECG signal. In the algorithm, the input waveform is linearly approximated by "split-and-merge" method, and then each line segment is symbolized with primitive set. The peeks in the symbolized input waveform are recognized by the finite-state automata, which the deterministic finite-state language is parsed by. This proposed algorithm correctly detects peaks in a normal ECG signal as well as in the abnormal ECG signal such as tachycardia and the contaminated signal with noise.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1993년도 추계학술대회
• /
• pp.137-140
• /
• 1993

This paper describes a Chaos analyzer and its applications to characteristic analysis of ECG signals and the other signals. We can detect chaotic system among the various system by quantitative and qualitative analysis using the proposed system. And we also propose a new Possibility to recognize abnormal state of ECG signal using the chaotic characteristcs of signal.

• 한국수정란이식학회:학술대회논문집
• /
• 한국수정란이식학회 2002년도 국제심포지엄
• /
• pp.73-73
• /
• 2002

Nuclear transfer (NT) techniques have advanced in the last years, and cloned animals have been produced by using somatic cells in several species including pig. However, it is difficult that the nuclear transfer porcine embryos development to blastocyst stage overcoming the cell block in vitro. Abnormal segregation of chromosomes in nuclear transferred embryos on genome activation stage bring about embryo degeneration, abnormal blastocyst, delayed and low embryo development. Thus, we are evaluated that the correlations of the frequency of embryo developmental rates and chromosome aberration in NT and In viかo fertilization (IVF) derived embryo. We are used for ear-skin-fibroblast cell in NT. If only karyotyping of embryonic cells are chromosomally abnormal, they may difficultly remain undetected. Then, we evaluate the chromosome aberrations, fluorescent in situ hybridization (FISH) with porcine chromosome 1 submetacentric specific DNA probe were excuted. In normal diploid cell nucleus, two hybridization signal was detected. In contrast, abnormal cell figured one or three over signals. The developmental rates of NT and IVF embryos were 55% vs 63%, 32% vs 33% and 13% vs 17% in 2 cell, 8 cell and blastocyst, respectively. When looking at the types of chromosome aberration, the detection of aneuploidy at Day 3 on the embryo culture. The percentage of chromosome aneuploidy of NT and IVF at 4-cell stage 40.0%, 31.3%, respectively. This result indicate that chromosomal abnormalities are associated with low developmental rate in porcine NT embryo. It is also suggest that abnormal porcine embryos produced by NT associated with lower implantation rate, increase abortion rate and production of abnormal fetuses.

• 대한임베디드공학회논문지
• /
• 제15권3호
• /
• pp.119-127
• /
• 2020

The detection and recognition of abnormal driving becomes crucial for achieving safety in Intelligent Transportation Systems (ITS). This paper presents a feature extraction method based on spectral data to train a neural network model for driving behavior recognition. The proposed method uses a two stage signal processing approach to derive time-saving and efficient feature vectors. For the first stage, the feature vector set is obtained by calculating variances from each frequency bin containing the power spectrum data. The feature set is further reduced in the second stage where an intersection method is used to select more significant features that are finally applied for training a neural network model. A stream of live signals are fed to the trained model which recognizes the abnormal driving behaviors. The driving behaviors considered in this study are weaving, sudden braking and normal driving. The effectiveness of the proposed method is demonstrated by comparing with existing methods, which are Particle Swarm Optimization (PSO) and Convolution Neural Network (CNN). The experiments show that the proposed approach achieves satisfactory results with less computational complexity.

• 인터넷정보학회논문지
• /
• 제20권6호
• /
• pp.137-142
• /
• 2019

본 논문은 이상 상황을 탐지하고 모니터링하는 다양한 서비스가 존재한다. 하지만 대부분의 서비스는 화재, 가스누출에 초점을 맞추어 진행되고 있으며, 독거노인과 중증장애인들의 사망 혹은 심정지 등 위급상황에 대하여 사전 예방 및 위급상황 대응이 불가능하다. 본 연구에서는 여러 생체신호 중 가장 위중하다고 판단되는 심박 신호의 이상 상태를 탐지하기 위하여 인공지능 모델을 설계하는 과정에서 적합한 데이터 변형과 모델을 비교한다. 세부적으로는 오픈 의료 데이터 PhysioNet의 MIT-BIH Arrhythmia Database를 이용하여 심전도(ECG) 데이터를 수집하고, 수집한 데이터를 각각 다른 방법으로 데이터를 변형한 후 학습하여 기본 심전도 데이터를 이용해 학습한 인공지능 모델과 비교한다.

• 한국해양공학회:학술대회논문집
• /
• 한국해양공학회 2004년도 학술대회지
• /
• pp.360-364
• /
• 2004

The objective of this study is to estimate the feasibility of acoustic emission method Jar the internal leak from the valves in nuclear power plants. The acoustic emission method was applied to the valves at the site, and the background noise was measured for the abnormal plant condition. From the comparison of background noise data with the experimental results as to relation between leak flow and acoustic signal, the minimum leak flow rates that am be detected by acoustic signal was suggested. When the background noise level are higher than the acoustic signal, the method described below was considered that the analysis the remainder among the background noise frequency spectrum and the acoustic signal spectrum.

• 한국멀티미디어학회논문지
• /
• 제19권11호
• /
• pp.1862-1870
• /
• 2016

A respiration is one of the most useful techniques for vital checking and an abnormal respiration is often the earliest sign of critical illness. Detection of respiration is based on the photo-plethysmography (PPG) with photodiode technique. Because PPG sensor using photodiode can be easily miniaturized, it is suitable for wearable devices. A system to measure respiration rate based on PPG signal is implemented and for a reliable measurement an improved algorithm in accuracy using PPG signal pattern is proposed in this paper. As results regarding to three types of respirations (regular interval, free interval, and weak respiration) the proposed algorithm showed error rate of 0.047, 0.067, and 0.122 respectively.

• 조명전기설비학회논문지
• /
• 제28권3호
• /
• pp.84-90
• /
• 2014

The partial discharge of Cast-resin Transformer has a difficulty to be analyzed, because it is an abnormal condition signal of which stochastic characteristics varies with time variance. In this study, background noise coming from the outside of the cast-resin transformers through ground wire can be removed and only a discharge signal of which defects are simulated can be obtained, using the wavelet transform method, which is a time-frequency domain analysis technique. As a result, it was confirmed that de-noising using the SWT technique is the best efficient among three methods of the wavelet transform techniques.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2007년도 춘계학술대회 논문집
• /
• pp.922-927
• /
• 2007

The present urban maglev which has been developed in Korea is controlled by 4-edge control method over each bogie. The control output which is derived from two gap sensors and one vertical acceleration sensor controls magnet to maintain a nominal gap. But, the gap signal acts as a big disturbance in rail joint though two gap sensors are used and finally result in unstable response and poor ride comfort. This paper treats of a method to compensate the gap signal in rail joint for the levitation control of urban maglev. The physically abnormal change of gap is detected when one gap sensor passes a rail joint, the disturbance of gap in rail joint is estimated. Finally the disturbance in gap signal is eliminated by processing the information of vehicle speed and estimated disturbance in when the other gap sensor passes a rail joint.