• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.99-101
• /
• 2022

이상 행동을 탐지하는 딥러닝 기반 검지 시스템은 동영상 기반 데이터로부터 움직임을 보이는 객체를 추적하고 그 객체의 행동을 분석하여 정상적인 행동 범위를 벗어나는 패턴을 보이는 영역을 이상으로 탐지한다. 특히 생성적 적대 신경망(GAN)과 광학 흐름 추정(Optical flow estimation) 기법을 활용하여 움직임에 대한 특징 정보를 추출하고 이를 학습하여 행동 패턴에 대한 모델링을 수행한다. 모델 학습 및 테스트에 활용되는 데이터셋의 해상도가 낮거나 이상 행동을 표현하는 특징 정보가 부족할 경우 최종 모델 성능에 부정적 영향을 미치게 되며, 특히 광학 흐름이 표현하는 이동량 측면에서 차이가 크게 나지 않는 이상 객체의 경우 탐지가 정확하게 이뤄지지 않는다. 본 연구에서는 동영상 프레임에서 나타나는 객체의 평균 종횡비를 구하고 정상적인 비율을 벗어나는 객체에 대해서 이상 행동을 취하는 샘플로 처리하는 후처리단 모듈을 제안하여 최종적인 모델 성능을 향상시키는 방법을 고안한다.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• v.16 no.4
• /
• pp.261-268
• /
• 2013

Purpose: To investigate the clinical characteristics and outcomes among infants and toddlers with failure to thrive (FTT). Methods: This retrospective study was done with 123 patients who had visited Pusan National University Children's Hospital during their first two years of life and had received an FTT diagnosis. We compared the clinical characteristics of the patients based on the causes of their FTT and their ages at the time of first hospital visit. We investigated triggering factors, feeding practices, and outcomes in 25 patients with nonorganic FTT (NOFTT). Results: Eighty cases (65.0%) were NOFTT. The gestational ages, birth weights, and weights at the first visits were significantly lower in patients with organic FTT (OFTT) (p<0.05). Infants who had first visited the clinic at age <6 months had the least z-score. The percentage of patients with severe weight decline was higher in OFTT than in NOFTT (60.0% vs. 17.3%). The z-scores at the follow-up visits were improved after treatment in both of the groups. Preceding infection was the most common triggering factor of NOFTT and persecutory feeding as abnormal behavior of caregiver was observed in 22 cases (88.0%). After treatment with feeding method modification, all patients with NOFTT showed normal growth. Conclusion: Weight decline is more severe in OFTT patients and in younger patients at the first visit. Infants with FTT can attain normal weight gain growth by treating organic diseases and supplying proper nutrition in OFTT, and by correcting abnormal dietary behavior of caregiver in NOFTT.

• Journal of Korea Multimedia Society
• /
• v.9 no.8
• /
• pp.1030-1036
• /
• 2006

The rapid proliferation of wireless networks and mobile computing applications has changed the landscape of network security. Anomaly detection is a pattern recognition task whose goal is to report the occurrence of abnormal or unknown behavior in a given system being monitored. This paper presents an efficient rough set based anomaly detection method that can effectively identify a group of especially harmful internal masqueraders in cellular mobile networks. Our scheme uses the trace data of wireless application layer by a user as feature value. Based on the feature values, the use pattern of a mobile's user can be captured by rough sets, and the abnormal behavior of the mobile can be also detected effectively by applying a roughness membership function considering weighted feature values. The performance of our scheme is evaluated by a simulation. Simulation results demonstrate that the anomalies are well detected by the method that assigns different weighted values to feature attributes depending on importance.

• Journal of the Korean Vacuum Society
• /
• v.7 no.3
• /
• pp.187-194
• /
• 1998

InGaAsP epitaxial layers lattice matched to InP were grown at 600 and $620^{\circ}C$ by low pressure metalorganic chemical vapor deposition. Solid phase composition of group III was controlled by the diffusion flux gas phase to the reachion surface. For the case of group V, the difference of As and P vapor pressure and pyrolysis efficiency of $PH_3$and $AsH_3$ mainly determined their in corporation into solid. An abnormal behavior of peak energy shift was observed below 75K in temperature variant photoluminescence study. This abnormal behavior was explained by the difference in order of ordering which makes spatial variation of energy gap in InGaAsP layer and this explanation was supported by the analyses of transmission electron microscopy and transmission spectroscopy.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.47-56
• /
• 2018

National-scale industrial control systems for gas, electric power, water processing, nuclear power, and traffic control systems increasingly use open networks and open standards protocols based on advanced information and communications technologies. The frequency of cyberattacks increases steadily because of the use of open networks and open standards protocols, but follow-up actions are limited. Therefore, the application of security solutions to an industrial control system is very important. However, it is not possible to apply security solutions to a real system because of the characteristics of industrial control systems. And a security system that can detect attacks without affecting the existing system is imperative. Therefore, in this paper, we propose an intrusion detection system based on packet analysis that can detect anomalous behaviors without affecting the industrial control system, and we verify the effectiveness of the proposed intrusion detection system by applying it in a test bed simulating a real environment.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.2
• /
• pp.223-233
• /
• 2013

At local emerging stock markets such as Korea, Hong Kong, Singapore and Taiwan, foreign investors (FI) are recognized as important investment community due to the globalization and deregulation of financial markets. Therefore, it is required to monitor the behavior of FI against a sudden enormous selling stocks for the concerned local governments or private and institutional investors. The main aim of this study is to propose an early warning system (EWS) which purposes issuing a warning signal against the possible massive selling stocks of FI at the market. For this, we suggest machine learning algorithm which predicts the behavior of FI by forecasting future conditions. This study is empirically done for the Korean stock market.

• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.10 no.2
• /
• pp.124-129
• /
• 1999

Background and Objectives : Classical singers are capable of masking abnormalities due to their high level of training and may present with apparent technical deficits rather than with obvious dysfunction. Therefore, some variations from expected normal laryngeal behavior may be present in trained classical singers. Consequently it is important for otolaryngologist to obtain a baseline assessment of their laryngeal function. Materials and Methods : Acoustic measurement including strobovideolaryngoscopy from 50 classically trained singers was done for this study, which was compared with the data from 20 untrained adults. Results and Conclusion : This study showed that 50-healthy asymptomatic classical singers revealed an incidence of 50% abnormal strobovideolaryngoscopic findings, but their acoustic data was within normal limit despite the abnormal laryngeal findings. Therefore the author recommends that the classical singers need objective voice analysis and their baseline data should be used for the accurate diagnosis of the cause of voice dysfunction In classical singer whose baseline laryngeal behavior may be unusual.

• Journal of Korean society of Dental Hygiene
• /
• v.15 no.6
• /
• pp.947-961
• /
• 2015

Objectives: The purpose of the study was to identify the relationship between metabolic syndrome and oral diseases in the middle aged and elderly in Korea. Methods: The study subjects were 6,390 people over 40 years old from 2010 and 2012 Korea National Health and Nutrition Examination Survey. The survey questionnaire consisted of health, nutrition, and oral examination surveys. The independent variables included general characteristics, health behavior, oral health behavior, and metabolic syndrome. The dependent variables included dental caries experience and periodontal disease. The oral examination was carried out by the dentist based on World Health Organization standard. Results: The average prevalence rate of metabolic syndrome MS was 23.79%, including 54.84% of risk group and 21.37% of normal group. The missing teeth rate was 82.38%, DMFT rate was 90.28% and the periodontal disease rate was 33.15%. Those having abnormal fasting blood glucose had 1.17 fold(95% CI: 1.00~1.37) higher periodontal disease than the normal group. The abnormal HDL cholesterol group had 1.25 times higher odds ratio(95% CI: 1.07~1.46) and the obese group had 1.27 times higher odds ratio(95% CI: 1.07~1.51). The risk group had 1.20 times higher odds ration(95% CI: 1.00~1.44) and that of the metabolic syndrome group was 1.60 times higher(95% CI: 1.29~1.97) in periodontal disease. The high blood pressure group had 1.25 times of missing teeth prevalence rate(95% CI: 1.00~1.37). The metabolic syndrome group had 1.47 times of missing teeth prevalence rate(95% CI: 1.11~1.94). Conclusions: The middle aged and elderly people in Korea had higher rate of metabolic syndrome and oral disease. It is necessary to implement the preventive oral health examination for the control of metabolic syndrome and oral diseases prevalence.

• Journal of Digital Contents Society
• /
• v.16 no.1
• /
• pp.25-32
• /
• 2015

With the increased frequency of crime such as assaults and sexual violence, the reliance on CCTV in arresting criminals has increased as well. However, CCTV, which should be monitored by human labor force at all times, has limits in terms of budget and man-power. Thereby, the interest in intelligent security system is growing nowadays. Expanding the techniques of an objects behavior recognition in previous studies, we propose a system to detect forms of violence between 2~3 objects from images obtained in CCTV. It perceives by detecting the object with the difference operation and the morphology of the background image. The determinant criteria to define violent behaviors are suggested. Moreover, provable decision metric values through measurements of the number of violent condition are derived. As a result of the experiments with the threshold values, showed more than 80% recognition success rate. A future research for abnormal behaviors recognition system in a crowded circumstance remains to be developed.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.20 no.4
• /
• pp.1-8
• /
• 2016

Recently, buildings tend to be large size, complex shape and functional. As the size of buildings is becoming massive, the need for structural health monitoring(SHM) technique is ever-increasing. Various SHM techniques have been studied for buildings which have different dynamic characteristics and are influenced by various external loads. Generally, the visual inspection and non-destructive test for an accessible point of structures are performed by experts. But nowadays, the system is required which is online measurement and detect risk elements automatically without blind spots on structures. In this study, in order to consider the response of non-linear structures, proposed a signal feature extraction and the adaptive threshold setting algorithm utilized to determine the abnormal behavior by using statistical methods such as control chart, root mean square deviation, generalized extremely distribution. And the performance of that was validated by using the acceleration response of structures during earthquakes measuring system of forced vibration tests and actual operation.