• 한국막학회:학술대회논문집
• /
• 한국막학회 2004년도 Proceedings of the second conference of aseanian membrane society
• /
• pp.203-206
• /
• 2004

Many types for environmental pollutant of endocrine disruptors have been reported on abnormal sexual development and abnormal feminizing responses of animals in a number of literatures [1]. Conventional biological methods for the removal of pollutants in wastewater require long times, and chemical oxiation methods in general cannot completely eliminate.(omitted)

• 한국컴퓨터정보학회논문지
• /
• 제26권4호
• /
• pp.11-19
• /
• 2021

이상 객체란 일반적이고 평범한 행동을 취하는 객체가 아닌 비정상적이고 흔하지 않은 행동을 하여 관찰이나 감시·감독을 필요로 하는 사람, 물체, 기계 장치 등을 뜻한다. 이를 사람의 지속적인 개입 없이 인공지능 알고리즘을 통해 탐지하기 위해서 광학 흐름 기법을 활용한 시간적 특징의 특이도를 관찰하는 방법이 많이 활용되고 있으며, 이 기법은 정해진 표현 범위가 없는 수많은 이상 행동을 식별하기에 적합하다. 본 연구에서는 생성적 적대 신경망(Generative Adversarial Network, GAN)으로 입력 영상 프레임을 광학 흐름 영상으로 변환하는 알고리즘을 학습시켜 비정상적인 상황을 식별한다. 특히 생성적 적대 신경망 모델이 입력 영상에 대한 중요한 특징 정보를 학습하고, 그 외 불필요한 이상치를 제외시키기 위한 전처리 과정과 학습 후 테스트 데이터셋에서 식별 정확도를 높이기 위한 후처리 과정을 고도화하여 전체적인 모델의 이상 행동 식별 성능을 향상시키는 기법을 제안한다. 이상 행동을 탐지하기 위한 학습 데이터셋으로 UCSD Pedestrian, UMN Unusual Crowd Activity를 활용하였으며, UCSD Ped2 데이터셋에서 프레임 레벨 AUC 0.9450, EER 0.1317의 수치를 보이며 이전 연구에서 도출된 성능 지표 대비 성능 향상이 확인되었다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2009년도 추계학술대회 논문집
• /
• pp.307-308
• /
• 2009

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2006년도 추계학술대회논문집
• /
• pp.556-561
• /
• 2006

In this study, the failure of the automotive clutch disk was investigated. During the process of power transmission, clutch disk plates did repeated work of releasing and engaging the pressure plate. The effects of unbalance rotation in the abnormal vibration and torque amplitude under engaged state were measured from this experiment. In order to reduce the unbalance, a modified clutch disk shape was developed. With a three-dimensional model of the stopper pin, to predict fatigue fracture, finite element analysis was carried out and evaluated the improvement of the new clutch disk.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.1046-1050
• /
• 2000

The necessity of diagnosis of the rotating machinery which is widely used in the industry is increasing. If rotating machinery has fault, we can detect fault using vibration or noise. But, in diagnosing rotating machinery, the end user who doesn't have expert knowledge needs the help of vibration diagnosis expert. However, vibration diagnosis experts who well satisfy the demand of end user are rare. So, this paper propose a development of the case-based reasoning system for abnormal vibration diagnosis of rotating machinery we construct the past experiences of vibration diagnosis expert into case base and shear the experiences of diagnosis expert with the end user. In this paper, we describe that process of structured system and adapting result of abnormal vibration diagnosis of electric motor.

• 산업융합연구
• /
• 제1권1호
• /
• pp.127-142
• /
• 2003

This Study examines the implications for event studies using the daily stock data. The output present the event study results. The event period is defined from 30 days before through 30 days after the event date, and is broken into four "windows" for abnormal return cumulation: the pre-event period, days -30 through -2; dajys -1 and 0, a period commonly investigated for the immediate impact of the event; and the post-event period, days +1 through +30. It shows how firm's information offerings affect the price process and consequent issues. The Patell Z test is an examples of a standardized abnormal return approach, which estimate a separate standard error for each security-event and assumes cross-sectional independence. The generalized sign test adjusts for the fraction of positive abnormal returns in the estimation period instead of assuming 0.5.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.1021-1024
• /
• 2001

Out of all metal-cutting processes, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. In this paper, the vision system of the sensing methods of drill flank wear on the basis of image processing is used to detect the wear pattern by non-contact and direct method and get the reliable wear information about drill. In image processing of acquired image, median filter is applied for noise removal. The vision flank wear area of the drill was measured. Backpropagation neural networks (BPns) were used for no-line detection of drill wear. The neural network consisted of three layers: input, hidden and output. The input vectors comprised of spindle rotational speed, feed rates, vision flank wear, thrust and torque signals. The output was the drill wear state which was either usable or failure. Drilling experiments with various spindle rotational speed and feed rates were carried out. The learning process was peformed effectively by utilizing backpropagation. The detection of the abnormal states using BPNs achieved 96.4% reliability even when the spindle rotational speed and feedrate were changed.

• 한국정밀공학회지
• /
• 제18권10호
• /
• pp.101-108
• /
• 2001

In this study, the static and dynamic characteristics of endmilling process were modelled and the analytic realization of chatter mechanism was discussed. In this reward, We have discussed on the comparative assessment of recursive time series modeling algorithms that cal represent time machining process and detect the abnormal machining behaviors in precision endmilling operation. In this study, simulation and experimental works were performed to show the malfunctional behaviors. For this purpose, new recursive algorithm(RLSM) was adopted for the oil-line system identification and monitoring of a machining process, we can apply these new algorithms in real process for detection of abnormal chatter. Also, The stability lobe of chatter was analysed by varying parameter of cutting dynamics in regenerative chatter mechanics.

• Transactions on Electrical and Electronic Materials
• /
• 제13권3호
• /
• pp.139-143
• /
• 2012

Optical emission spectroscopy (OES) is often used for real-time analysis of the plasma processes. OES has been suggested as a primary plasma process monitoring tool. It has the advantage of non-invasive in-situ monitoring capability but selecting the proper wavelengths for the analysis of OES data generally relies on empirically established methods. In this paper, we propose a practical method for the selection of OES wavelength peaks for the analysis of plasma etch process and this is done by investigating reactants and by-product gas species that reside in the plasma etch chamber. Wavelength selection criteria are based on the standard deviation and correlation coefficients. Moreover, chemical actinometry is employed for the normalization of the selected wavelengths. We also present the importance of chemical actinometry of OES data for quantitative analysis of plasma. Then, the suggested OES peak selection method is employed.. This method is used to find out the reason behind abnormal etching of PR erosion during a series of $SiO_2$ etch processes using the same recipe. From the experimental verification, we convinced that OES is not only capable for real-time detection of abnormal plasma process but it is also useful for the analysis of suspicious plasma behavior.

• 제어로봇시스템학회논문지
• /
• 제6권6호
• /
• pp.486-491
• /
• 2000

The dicing process cuts a semiconductor wafer to lengthwise and crosswise direction by using a rotating circular diamond blade. But inferior goods are made under the influence of several parameters in dicing such as blade, wafer, cutting water and cutting conditions. This paper describes a monitoring algorithm using neural network in order to find out an instant of vibration signal change when bad dicing appears. The algorithm is composed of two steps: feature extraction and decision. In the feature extraction, five features processed from vibration signal which is acquired by accelerometer attached on blade head are proposed. In the decision, back-propagation neural network is adopted to classify the dicing process into normal and abnormal dicing, and normal and damaged blade. Experiments have been performed for GaAs semiconductor wafer in the case of normal/abnormal dicing and normal/damaged blade. Based upon observation of the experimental results, the proposed scheme shown has a good accuracy of classification performance by which the inferior goods decreased from 35.2% to 6.5%.