• 한국신뢰성학회지:신뢰성응용연구
• /
• 제18권1호
• /
• pp.20-32
• /
• 2018

Purpose: The purpose of this study is to set up an anomaly detection criteria for sensor data coming from a motorcycle. Methods: Five sensor values for accelerator pedal, engine rpm, transmission rpm, gear and speed are obtained every 0.02 second from a motorcycle. Exploratory data analysis is used to find any pattern in the data. Traditional process control methods such as X control chart and time series models are fitted to find any anomaly behavior in the data. Finally unsupervised learning algorithm such as k-means clustering is used to find any anomaly spot in the sensor data. Results: According to exploratory data analysis, the distribution of accelerator pedal sensor values is very much skewed to the left. The motorcycle seemed to have been driven in a city at speed less than 45 kilometers per hour. Traditional process control charts such as X control chart fail due to severe autocorrelation in each sensor data. However, ARIMA model found three abnormal points where they are beyond 2 sigma limits in the control chart. We applied a copula based Markov chain to perform statistical process control for correlated observations. Copula based Markov model found anomaly behavior in the similar places as ARIMA model. In an unsupervised learning algorithm, large sensor values get subdivided into two, three, and four disjoint regions. So extreme sensor values are the ones that need to be tracked down for any sign of anomaly behavior in the sensor values. Conclusion: Exploratory data analysis is useful to find any pattern in the sensor data. Process control chart using ARIMA and Joe's copula based Markov model also give warnings near similar places in the data. Unsupervised learning algorithm shows us that the extreme sensor values are the ones that need to be tracked down for any sign of anomaly behavior.

• 한국컴퓨터정보학회논문지
• /
• 제11권5호
• /
• pp.157-164
• /
• 2006

최근에는 대부분의 인터넷 공격은 악성코드(Malware)에 의한 잘 알려지지 않은 제로데이 공격 형태가 주류를 이루고 있으며, 이미 알려진 공격유형들에 대해서 탐지하는 오용탐지 기술로는 이러한 공격에 대응하기가 어려운 실정이다. 또한, 다양한 공격 패턴들이 인터넷상에 나타나고 있기 때문에 기존의 정보 보호 기술로는 한계에 다다르게 되었고, 웹기반 서비스가 보편화됨에 따라 인터넷상에 노출된 웹 서비스가 주공격 대상이 되고 있다. 본 논문은 인터넷상의 트래픽 유형을 분류하고, 각 유형에 따른 이상 징후를 탐지하고 분석할 수 있는 비정상행위공격 탐지기술(Anomaly Intrusion Detection Technologies)을 포함하고 있는 위협관리 시스템(Threat Management System)을 제안한다.

• 한국수소및신에너지학회논문집
• /
• 제33권1호
• /
• pp.85-94
• /
• 2022

To detect anomaly signs of thermal runaway in advance, this study analyzed the signals from various sensors installed in lithium-ion batteries. The thermal runaway mechanism was analyzed, and measurement variables for anomalies of a battery cell were surface temperature, strain, and gas concentration. The changes and characteristics of three variables during the thermal runaway process were analyzed under the abuse environment: the overheat and the overcharge. In experiment, the thermal runaway of the battery proceeded in the initial developing stage, the outgassing stage, and the ignition stage. Analysis from the measured data indicated that the suitable variable to detect all stages of thermal runaway is the surface temperature of the battery, and surface strain is alternative.

• 정보처리학회논문지C
• /
• 제11C권6호
• /
• pp.719-724
• /
• 2004

현재는 인터넷 이용자들이 급격하게 증가하고 있으며, 초보수준의 일반 네트워크 사용자들도 인터넷상의 공개된 해킹 도구들을 사용하여 고도의 기술을 요하는 침임이 가능해졌기 때문에 해킹 문제가 더욱 심각해지고 있다. 해커들이 침입하기 위하여 취약점을 알아내려고 의도하는 다양한 형태의 침입시도들을 탐지하여 침입이 일어나는 것을 사전에 방어할 수 있는 침입시도탐지가 적극적인 예방 차원에서 더욱 필요하다. 기존의 포트 스캔이나 네트워크 취약점 검색 공격에 대응하기 위한 네트워크 기반의 비정상 침입시도 탐지 알고리즘들은 침입시도탐지에 있어 몇 가지 한계점을 갖고 있다. 기존 알고리즘들의 취약성은 Slow Scan과 Coordinated Scan을 한 경우 탐지한 수 없다. 따라서, 침입시도 유형에 제한을 받지 않고 침입시도에 관한 다양한 형태의 비정상 접속을 효과적으로 탐지할 수 있는 새로운 개념의 알고리즘이 요구된다. 본 논문에서는 평상시 정상적인 서비스 패턴을 가지고 그 패턴과 다른 비정상 서비스 패턴이 보이면 이를 침입시도로 탐지하는 개념의 SPAD(Session Pattern Anomaly Detector) 기법을 제안한다.

• Advances in pediatric surgery
• /
• 제8권2호
• /
• pp.107-112
• /
• 2002

Branchial anomaly is a frequently occurring congenital abnormality in childhood. It is important for the pediatric surgeon alike to be familiar with the embryology and differentiation of head and neck structure to accurately diagnose and treat these lesions. Eighty-five patients with branchial anomaly treated at Hanyang University Hospital between 1980 and 2001 were reviewed to determine relative frequency, clinical classification and appropriate treatment. The male to female ratio of branchial anomaly was 1.2:1. The most commonly presenting age was before 1 year (32%) and the age group between 1 and 3 year (22%) followed it. According to the classification of branchial anomalies, 73 of 85 cases were second branchial anomaly, 9 had the first type and 3 did fourth type. One patient showed combined anomalies of the first and the second type. Infection sign were seen in 70% of patients at the time of the first visit to our hospital and also patients' symptoms were frequently related with the infection. Forty-one cases (48%) were fistula, 21 (25%) were cysts, 21 (25%) were sinuses, and two were only cartilage remnants. The most common type of the branchial anomalies is the second branchial fistula and the most common symptoms of the anomalies are related with infection. Initial proper diagnosis and anatomical classification of the anomalies are very important in managing the lesions. The efforts to find the exact anatomical location of the fistula or sinus tract are necessary because total excision of the lesions including those tracts is the only way to prevent recurrence.

• 인터넷정보학회논문지
• /
• 제24권1호
• /
• pp.49-59
• /
• 2023

COVID-19로 대표되는 팬데믹 상황에서 의료 인력 부족으로 인한 문제가 대두되고 있다. 본 논문에서는 진단 업무를 지원하기 위한 컴퓨터 비전 솔루션으로 PA 흉부 X-선 영상에 대한 병변 유무 진단 방법에 대해 제시한다. 디지털 영상에 대한 특징 비교 방식의 이상 탐지 기법을 X-선 영상에 적용하여 비정상적인 영역을 예측할 수 있다. 정렬된 PA 흉부 X-선 영상으로부터 특징 벡터를 추출하고 패치 단위로 분할하여 지역적으로 등장하는 비정상을 포착한다. 사전 실험으로 다중 객체를 포함하는 시뮬레이션 데이터 세트를 생성하고 이에 대한 비교 실험 결과를 제시한다. 정렬된 영상에 대해 적용 가능한 패치 특징 하드마스킹을 통해 프로세스의 효율성 및 성능을 향상하는 방법을 제시한다. 지역 특수성 및 전역 이상 탐지 결과를 합산하여 기존 연구 대비 6.9%p AUROC 향상된 성능을 보인다.

• Advances in pediatric surgery
• /
• 제17권2호
• /
• pp.162-169
• /
• 2011

Branchial cleft anomalies are the second most common head and neck congenital lesions seen in children. Amongst the branchial cleft malformations, second cleft lesions account for 95 % of the branchial anomalies. This article analyzes all the cases of second branchial cleft anomalies operated on at Seoul National University Hospital from September 1995 to February 2011. We analyzed sex, age, symptom and sign, accompanied anomaly, diagnosis, treatment, pathologic report and outcome via retrospective review of medical records. In this series, we had 61 patients (27 female and 34 male). The mean age at the time of operation was 38 months. 31 lesions were on the right, 20 were on the left and 10 were bilateral. The most frequent chief complaints at presentation were non-tender mass and cervical opening without any discharge. According to anatomic type, 29 patients had branchial cleft sinuses, 14 had cysts, 14 had fistulas and 4 had skin tags. Complete excision was attempted if possible and antibiotics challenged when infection was suspected. Complete excision was achieved in 96.7 % of cases. Incision and drainage was done in 2 cases due to severe inflammation, and both recurred. Postoperative complications included wound infection in 2 cases. Microscopic examonation revealed squamous epithelium in 90.2 % and squamous metaplasia in one case in the branchial cleft cyst wall. In summary, second branchial anomaly is found more frequently on right side of neck. Fistulas are diagnosed earlier than cystic forms. Most cases could be diagnosed by physical examination. The definitive treatment is complete excision and sufficient antibiotics coverage for cases with inflammation. After drainage of infected lesions, follow up excision after 1 year might be beneficial for preventing recurrence.

• 한국컴퓨터정보학회논문지
• /
• 제10권6호
• /
• pp.9-16
• /
• 2005

최근 인터넷 이용자들이 급격하게 증가하고 있으며, 초보수준의 일반 네트워크 사용자들도 인터넷상의 공개된 해킹 도구들을 사용하여 고도의 기술을 요하는 침입이 가능하여 해킹 문제가 더욱 심각해지고 있다. 해커들이 침입하기 위하여 취약점을 알아내려고 의도하는 다양한 형태의 침입시도를 사전에 탐지하여 침입이 일어나는 것을 미리 방어할 수 있는 침입시도탐지가 적극적인 예방 차원에서 더욱 필요하다. 기존의 포트 스캔이나 네트워크 취약점 공격에 대응하기 위한 네트워크 기반의 비정상 침입시도 탐지 알고리즘은 침입시도함지에 있어 몇 가지 한계점을 갖고 있다. 기존 알고리즘은 Slow Scan, Coordinated Scan을 할 경우 탐지할 수 없다는 것이다. 따라서 침입시도 유형에 제한을 받지 않고 침입시도에 관한 다양한 형태의 비정상 접속을 효과적으로 탐지할 수 있는 새로운 개념의 알고리즘이 요구된다. 본 논문에서는 세션 패턴과 탐지 오류율을 규칙기반으로 하는 침입시도 탐지알고리즘(Session patterns & FCM Anomaly Detector : SFAD)을 제안한다.

• Journal of Chest Surgery
• /
• 제13권4호
• /
• pp.448-454
• /
• 1980

The juxtaposition of atrial appendage may possibly be diagnosed by angiography recently, and it is widely regarded as an ominous sign of severe cyanotic congenital heart disease. However, it is apparently rare congenital cardiac anomaly in which the atrial appendages lie side by side, both to the left or right of the great arteries, known as left or right juxtaposition of the atrial appendages. Juxtaposition of the atrial appendages has no functional significance, since it does not, itself, cause any hemodynamic disturbance. But it`s presence always indicates the coexistence of other major cardiac anomalies. In review of literatures TGA and VSD are invariable present, and ASD is common. Other anomalies, such as, tricuspid atresia, pulmonary outflow that, obstruction bicuspid pulmonic valve, persistent SVC etc. are relatively high incidence. In this report, we present one case of 6 year old female child having left juxtaposition of atrial appendage combined with TGA [D-looping, D-transposition], TAPVD, large ASD, small VSD, and vertical vein.

• Progress in Superconductivity
• /
• 제4권1호
• /
• pp.32-36
• /
• 2002

We have measured the Hall effect in infinite-layer Sr／sub 0.9／La／sub 0.1／$CuO_2$ thin films grown by molecular beam epitaxy. We do not observe $T^{2}$ dependence of the cotangent of Hall angle, which is commonly observed in other cuprate High-Tc superconductors. Therefore, this result cannot be interpreted within two different scattering mechanism based on charge-spin separation theory. The mixed-state Hall effect shows no sign anomaly, implying that tile hydrodynamic contribution of vortex core is negligibly small.