• 대한산업공학회지
• /
• 제41권6호
• /
• pp.540-550
• /
• 2015

There have been many attempts to find knowledge from data using conventional statistics, data mining, artificial intelligence, machine learning and pattern recognition. In those research areas, knowledge is approached in two ways. Firstly, researchers discover knowledge represented in general features for universal recognition, and secondly, they discover exceptional and distinctive features. In process mining, an instance is sequential information bounded by case ID, known as process instance. Here, an exceptional process instance can cause a problem in the analysis and discovery algorithm. Hence, in this paper we develop a method to detect the knowledge of exceptional and distinctive features when performing process mining. We propose a method for anomaly detection named Distance-based Anomaly Process Instance Detection (DAPID) which utilizes distance between process instances. DAPID contributes to a discovery of distinctive characteristic of process instance. For verifying the suggested methodology, we discovered characteristics of exceptional situations from log data. Additionally, we experiment on real data from a domestic port terminal to demonstrate our proposed methodology.

• 정보과학회 논문지
• /
• 제42권4호
• /
• pp.483-486
• /
• 2015

얼굴 인식기반의 사용자 보안 시스템은 접근이 허가된 사용자의 사진이나 비디오를 이용한 공격에 취약하다는 단점을 가지고 있다. 본 연구에서는 인증되지 않은 사용자가 비디오를 이용하여 시스템에 접근할 경우 해당 공격 시도를 검출하기 위한 위변조(Spoof) 검출 방법을 제안한다. 제안하는 방법은 연속된 3개의 Frame에서 푸리에 변환과 Dense-SIFT 구분자를 사용하여 400개의 실제 및 위변조 비디오 영상을 대상으로 실험한 결과 99%의 검출 정확도를 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권8호
• /
• pp.2928-2947
• /
• 2015

Adaptive Gaussian filter (AGF) is a recently developed switching filter to remove salt and pepper noise. AGF first directly identifies pixels of gray levels 0 and 255 as noise pixels, and then only restored noise pixels using a Gaussian filter with adaptive variance based on the estimated noise density. AGF usually achieves better denoising effect in comparison with other filters. However, AGF still fails to obtain good denoising effect on images with noise-free pixels of gray levels 0 and 255, due to its severe false alarm in its noise detection stage. To alleviate this issue, a modified version of AGF is proposed in this paper. Specifically, the proposed filter first performs noise detection via an image block based noise density estimation and sequential noise density guided rectification on the noise detection result of AGF. Then, a modified Gaussian filter with adaptive variance and window size is used to restore the detected noise pixels. The proposed filter has been extensively evaluated on two representative grayscale images and the Berkeley image dataset BSDS300 with 300 images. Experimental results showed that the proposed filter achieved better denoising effect over the state-of-the-art filters, especially on images with noise-free pixels of gray levels 0 and 255.

• Journal of the Korean Data and Information Science Society
• /
• 제22권2호
• /
• pp.197-206
• /
• 2011

신호처리 관련 응용문제에서는 신호에서 실시간으로 발생하는 비정상적인 사건들을 탐지하는 것이 매우 중요하다. 이전에 알려져 있는 비정상 사건 탐지방법들은 신호에 대한 명확한 통계적인 모형을 가정하고, 비정상적인 신호들은 통계적인 모형의 가정 하에서 비정상적인 사건들로 해석한다. 탐지방법으로 최대우도와 베이즈 추정 이론이 많이 사용되고 있다. 그러나 앞에서 언급한 방법으로는 로버스트 하고 다루기 쉬운 모형을 추정한다는 것은 쉽지가 않다. 좀 더 로버스트한 모형을 추정할 수 있는 방법이 필요하다. 본 논문에서는 로버스트 하다고 알려져 있는 서포트 벡터 기계를 이용하여 온라인으로 비정상적인 신호를 탐지하는 방법을 제안한다.

• 한국군사과학기술학회지
• /
• 제21권2호
• /
• pp.225-234
• /
• 2018

In this paper, we consider a problem of partitioning a search area into smaller rectangular regions, so that multiple platforms can conduct search operations independently without requiring unnecessary coordination among themselves. The search area consists of cells where each cell has some prior information regarding the probability of target existence. The detection probability in particular cell is evaluated by multiplying the observation probability of the platform and the target existence probability in that cell. The total detection probability within the search area is defined as the cumulative detection probability for each cell. However, since this search area partitioning problem is NP-Hard, we decompose the problem into three sequential phases to solve this computationally intractable problem. Additionally, we discuss a special case of this problem, which can provide an optimal analytic solution. We also examine the performance of the proposed approach by comparing our results with the optimal analytic solution.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권1호
• /
• pp.58-73
• /
• 2014

As is well known, cooperative sensing can significantly improve the sensing accuracy as compared to local sensing in cognitive radio networks (CRNs). However, a large number of cooperative secondary users (SUs) reporting their local detection results to the fusion center (FC) would cause much overhead, such as sensing delay and energy consumption. In this paper, we propose a fast cooperative sensing scheme, called double threshold fusion (DTF), to reduce the sensing overhead while satisfying a given sensing accuracy requirement. In DTF, FC respectively compares the number of successfully received local decisions and that of failed receptions with two different thresholds to make a final decision in each reporting sub-slot during a sensing process, where cooperative SUs sequentially report their local decisions in a selective fashion to reduce the reporting overhead. By jointly considering sequential detection and selective reporting techniques in DTF, the overhead of cooperative sensing can be significantly reduced. Besides, we study the performance optimization problems with different objectives for DTF and develop three optimum fusion rules accordingly. Simulation results reveal that DTF shows evident performance gains over an existing scheme.

• 정보보호학회논문지
• /
• 제15권2호
• /
• pp.13-22
• /
• 2005

본 논문에서는 시스템 호출을 이용하여 이상 침입 탐지 시스템의 성능을 향상시키기 위해, 특징 선택과 가변 길이 데이터를 고정 길이 학습 패턴으로 변환 생성하는 문제를 해결하기 위한 사운덱스 알고리즘을 적용한 신경망 학습을 통하여 이상 침입 탐지의 연구를 하고자 한다. 즉, 가변 길이의 순차적인 시스템 호출 데이터를 사운덱스 알고리즘에 의한 고정 길이의 행위 패턴을 생성하여 역전파 알고리즘과 퍼지 멤버쉽 함수에 의해 신경망 학습을 수행하였다. 역전파 신경망과 뉴로-퍼지 기법을 UNM의 Sendmail Data Set을 이용하여 시스템 호출의 이상침입 탐지에 적용하여 시간과 공간 복잡도 그리고 MDL 측면에서 성능을 검증하였다.

• 한국자동차공학회논문집
• /
• 제6권3호
• /
• pp.115-122
• /
• 1998

For digital engine control timings, such as ignition, are based on the crank shaft angle. Therefore, it is very important that the angle of the crank shaft can be detected with accuracy for optimal ignition timing. Sequential multi-point injection(MPI) systems that have independent injection events for each cylinder, are used to inject an accurate quantity of fuel, and to cope with varying engine status promptly. In this study the distributorless ignition timing. A crankshaft position sensor has been installed such that it generates a number of pulses per crankshaft revolution to permit accurate detection of the crank shaft angle. An event detecting algorithm has been developed, which detects the crank shaft pulses generated by the position sensor, and the software outputs the required control signals at given crank angle values. We clarified that the hardware method is the best way to increase the performance of the control system, because the event detecting duration T(1+2)max becomes zero.

• 로봇학회논문지
• /
• 제7권2호
• /
• pp.92-100
• /
• 2012

A map of complex environment can be generated using a robot carrying sensors. However, representation of environments directly using the integration of sensor data tells only spatial existence. In order to execute high-level applications, robots need semantic knowledge of the environments. This research investigates the design of a system for recognizing objects in 3D point clouds of urban environments. The proposed system is decomposed into five steps: sequential LIDAR scan, point classification, ground detection and elimination, segmentation, and object classification. This method could classify the various objects in urban environment, such as cars, trees, buildings, posts, etc. The simple methods minimizing time-consuming process are developed to guarantee real-time performance and to perform data classification on-the-fly as data is being acquired. To evaluate performance of the proposed methods, computation time and recognition rate are analyzed. Experimental results demonstrate that the proposed algorithm has efficiency in fast understanding the semantic knowledge of a dynamic urban environment.

• 전자공학회논문지SC
• /
• 제48권3호
• /
• pp.40-47
• /
• 2011

본 논문에서는 상태 피드백 제어를 이용한 비동기 순차 머신의 고장 탐지 및 극복 과정을 다룬다. 논문에서 고려하는 비동기 머신은 간헐 고장의 영향을 받는다. 간헐 고장이 발생하면 머신은 원하지 않는 상태 천이를 하며, 일정 시간 동안 외부 입력의 변화에 반응하지 못하고 고장 상태를 유지한다. 본 논문에서는 비동기 머신에서 발생하는 간헐 고장을 탐지할 수 있는 조건을 규명하고 간헐 고장을 극복하는 견실한 상태 피드백 제어기의 존재조건과 설계 알고리듬을 제안한다. 과도 고장에 대한 기존의 내고장성 제어 기법과 이번 연구에서 제안하는 간헐 고장에 대한 제어기 동작과의 차이점도 설명한다. 또한 사례 연구를 통해서 제안된 제어기의 설계 과정을 예시한다.