• 대한임베디드공학회논문지
• /
• 제9권5호
• /
• pp.261-268
• /
• 2014

Face detection is essential to the full automation of face image processing application system such as face recognition, facial expression recognition, age estimation and gender identification. It is found that local image features which includes Haar-like, LBP, and MCT and the Adaboost algorithm for classifier combination are very effective for real time face detection. In this paper, we present a face detection method using local pixel direction code(PDC) feature and lookup table classifiers. The proposed PDC feature is much more effective to dectect the faces than the existing local binary structural features such as MCT and LBP. We found that our method's classification rate as well as detection rate under equal false positive rate are higher than conventional one.

• 한국지능시스템학회논문지
• /
• 제11권5호
• /
• pp.400-405
• /
• 2001

This paper addresses an effective approach of training neural networks classifiers for credit card fraud detection. The proposed approach uses evolutionary programming to trails the neural networks classifiers based on maximization of the detection rate of fraudulent usages on some ranges of the rejection rate, loot minimization of mean square error(MSE) that Is a common criterion for neural networks learning. This approach enables us to get classifier of satisfactory performance and to offer a directive method of handling various conditions and performance measures that are required for real fraud detection applications in the classifier training step. The experimental results on "real"credit card transaction data indicate that the proposed classifiers produces classifiers of high quality in terms of a relative profit as well as detection rate and efficiency.

• 반도체디스플레이기술학회지
• /
• 제12권2호
• /
• pp.57-61
• /
• 2013

This paper presents adaptive neuro-fuzzy inference based defect detection method for various defect types, such as micro-crack, fingerprint and contamination, in heterogeneously textured surface of polycrystalline solar wafers. Polycrystalline solar wafer consists of various crystals so the surface of solar wafer shows heterogeneously textures. Because of this property the visual inspection of defects is very difficult. In the proposed method, we use local binary feature and fuzzy reasoning for defect detection. Experimental results show that our proposed method achieves a detection rate of 80%~100%, a missing rate of 0%~20% and an over detection (overkill) rate of 9%~21%.

• 전기학회논문지
• /
• 제61권9호
• /
• pp.1358-1361
• /
• 2012

This paper presents a study on the analysis for activation time and threshold value of heat detection sensor using HRR(Heat Release Rate). And it is represented to quantity of heat to activate heat detection sensor. The experiment is conducted to measure activation time and HRR of fire detection sensor burning alcohol and n-heptane. In order to burn the alcohol and n-heptane using $43.5cm(L){\times}43.5cm(W){\times}5cm(D)$ and $33cm(L){\times}33cm(L){\times}5cm(D)$ steel pan and the quantity of alcohol and n-heptane are 2.5 L and 650 g, respectively. The results show that peak HRR are in case of alcohol 66.13 kW and in case of n-heptane 151.64 kW, respectively. Total heat releases of heat detection sensor are in case of alcohol approximately 20.7 MJ and in case of n-heptane approximately 18 MJ, respectively.

• 한국컴퓨터정보학회논문지
• /
• 제15권2호
• /
• pp.47-53
• /
• 2010

본 논문은 동영상에서 실시간 얼굴검출을 위하여 Residual Image 검출과 색상정보를 이용한 얼굴검출 기법을 제안하였다. 제안된 기법은 동영상에서 빠른 처리 속도와 높은 얼굴 검출율을 나타냈으며 기울어진 얼굴영상에 대한 보정작업을 통하여 검출 에러율을 줄였다. 실시간으로 전송된 동영상에서 검출의 대상이 되는 정지영상을 추출한다. 추출된 영상은 기울어진 얼굴검출을 위한 window회전 알고리즘을 사용하고 이렇게 보정된 영상은 얼굴 검출에 필요한 특징을 추출하기 위해 AdaBoost알고리즘을 사용하여 실시간으로 얼굴이 검출된 영상을 획득하게 된다.

• 한국정보통신학회논문지
• /
• 제17권6호
• /
• pp.1462-1470
• /
• 2013

본 논문에서는 클러스터 단위로 센서 노드의 키와 그룹키를 생성한 후 분배하는 KGDC(Key Generation and Distribution based on Cluster) 모듈과 이 키를 이용하여 시빌 공격과 웜홀 공격을 탐지할 수 있는 SWAD(Sybil & Wormhole Attack Detection) 모듈로 구성되는 SWAD-KNH(Sybil & Wormhole Attack Detection using Key, Neighbor list and Hop count) 기법을 제안한다. KGDC 모듈은 ECDH 알고리즘, 해시함수, 키 체인 기법을 이용하여 그룹키와 센서 노드의 키들을 생성하고 안전하게 배포한다. SWAD 모듈은 2단계 키 확인절차를 수행함으로써 시빌 공격 탐지를 강화시키고, 근원지와 목적지 노드의 공동이웃노드들의 수와 홉 카운트를 이용하여 웜홀 공격을 탐지한다. 그 결과 SWAD-KNH 기법은 시빌 공격 탐지율은 91.2%, 평균 FPR(False Positive Rate)은 3.82%이고, 그리고 웜홀 공격 탐지율은 90%, 평균 FPR(False Positive Rate)은 4.64%으로 평가되므로 시빌과 웜홀 공격 탐지율 및 시빌과 웜홀 공격 탐지의 신뢰성을 향상시켰다.

• 융합보안논문지
• /
• 제22권5호
• /
• pp.145-154
• /
• 2022

최근 봇넷의 광범위한 확산과 고도의 정교함은 기업과 사용자뿐만 아니라 국가 간 사이버전에도 심각한 결과를 초래하고 있다. 이 때문에 봇넷을 탐지하고자 하는 연구는 꾸준히 되고 있다. 하지만, DGA 기반의 봇넷은 기존의 시그니처 및 통계 기반의 기술로는 탐지율은 높지만, 오탐율 또한 높은 한계가 있다. 이에 본 논문에서는 DGA 기반의 봇넷을 탐지하고자 문자 기반의 n-gram을 활용한 탐지모델을 제안한다. 제안한 모델을 통해 기존의 탐지 기술의 한계인 탐지율을 높이고 오탐율을 최소화할 수 있다. 다양한 DGA 봇넷에서 사용하는 대규모의 도메인 데이터셋과 정상 도메인에 대한 실험을 통해 기존의 모델보다 성능이 우수함을 확인하였다. 제안된 모델의 오탐율은 2~4% 미만이며 전체 탐지 정확도와 F1 점수는 모두 97.5%임을 확인하였다. 이처럼 본 논문에서 제안한 모델을 통해 DGA 기반의 봇넷에 대한 탐지 및 대응 능력이 향상될 것을 기대한다.

• Parasites, Hosts and Diseases
• /
• 제23권2호
• /
• pp.285-292
• /
• 1985

The present study has been designed as a basic study on laboratory diagnosis of giardiasis and to demonstrate a more effective method for the detection of Giardia lamblia cyst with the inherent advantages of minimizing both the number of stool examinations required and the interval of stool collections for estimating the real state of prevalence in the shortest time possible. There were 3 subject groups of 75 children each currently residing in an orphanage in Gunsan city, Jeonbuk province from which stool specimens were collected every day, every other day, and every 3 days. The procedure is as follows: 1) resuspend the InKed sample after fixation with SAF solution 2) centrifuge the sediment for 1 min. at 2, 000 rpm after straining through gauze into a tube 3) divide the sediment into 3 parts and use them for direct fecal smear, formalin-ether concentration (MGL) and zinc sulfate ($ZnSO_4$) floatation techniques. The results are summarized as follows: 1. Overall infection rate after 10 trials showed a 60% positive indication. The positive rate among children under 4 years old was significantly higher than the rate in children over 4 years old. No significant difference in rate by sex was observed. 2. The results of eBaminations by direct fecal smear and MGL techniques appeared more accurate than that obtained by $ZnSO_4$ ftoatation method as indicated by a higher positive rate. Of all three methods concerned, combinations of two demonstrated a higher positive rate than that shown by any one alone. 3. In three consecutive examinations under varying conditions such as different days, the cyst detection rate by MGL technique indicated 83%. In 5 examinations under the same varying conditions, the indicated rate was 94%. 4. The interval of stool collection proved to be insignificant for the cyst detection rate. In conclusion, both MGL method and modified fecal direct smear can provide a good cyst detection rate of G. lamblia provided that more than 3 consecutive examinations of stool under varying conditions are carried out.

• KNOM Review
• /
• 제23권1호
• /
• pp.10-17
• /
• 2020

최근 보안이 취약한 IoT 장치를 악용하는 분산 서비스 거부 공격의 위협이 확산됨에 따라 신속하게 공격을 탐지하고 공격자의 위치를 찾기 위해 소스측 서비스 거부 공격 탐지 연구가 활성화되고 있다. 또한, 소스측 탐지의 지역적 한계를 극복하기 위해 개별 사이트에 위치한 소스측 네트워크들의 탐지 결과를 공유하는 협력형 소스측 공격 탐지 기법도 활성화되고 있다. 이 논문에서는 통계적 가중치를 이용하는 협력형 소스측 분산 서비스 거부 공격 탐지 기법의 성능을 평가한다. 통계적 가중치는 개별 소스측 네트워크의 시간대에 해당하는 탐지율과 오탐지율을 기반으로 계산된다. 제안된 기법은 여러 지역에서 발생한 소스측 서비스 거부 공격 탐지 결과들을 수집하고 가중치를 부여하여 결과를 도출하고, 이를 통해 DDoS 공격 발생 여부를 결정한다. 실제 DNS 요청 트래픽을 기반으로 실험한 결과, 제안된 기법은 높은 공격탐지율을 유지하면서, 공격오탐율을 2% 줄일 수 있음을 확인하였다.

• Journal of Computing Science and Engineering
• /
• 제5권4호
• /
• pp.305-313
• /
• 2011

Due to the expansion of high-speed Internet access, the need for secure and reliable networks has become more critical. The sophistication of network attacks, as well as their severity, has also increased recently. As such, more and more organizations are becoming vulnerable to attack. The aim of this research is to classify network attacks using neural networks (NN), which leads to a higher detection rate and a lower false alarm rate in a shorter time. This paper focuses on two classification types: a single class (normal, or attack), and a multi class (normal, DoS, PRB, R2L, U2R), where the category of attack is also detected by the NN. Extensive analysis is conducted in order to assess the translation of symbolic data, partitioning of the training data and the complexity of the architecture. This paper investigates two engines; the first engine is the back-propagation neural network intrusion detection system (BPNNIDS) and the second engine is the radial basis function neural network intrusion detection system (BPNNIDS). The two engines proposed in this paper are tested against traditional and other machine learning algorithms using a common dataset: the DARPA 98 KDD99 benchmark dataset from International Knowledge Discovery and Data Mining Tools. BPNNIDS shows a superior response compared to the other techniques reported in literature especially in terms of response time, detection rate and false positive rate.