• 제목/요약/키워드: False-pick

검색결과 6건 처리시간 0.026초

Alpha-cut과 Beta-pick를 이용한 시그너쳐 기반 침입탐지 시스템과 기계학습 기반 침입탐지 시스템의 결합 (A Combination of Signature-based IDS and Machine Learning-based IDS using Alpha-cut and Beta pick)

  • 원일용;송두헌;이창훈
    • 정보처리학회논문지C
    • /
    • 제12C권4호
    • /
    • pp.609-616
    • /
    • 2005
  • 시그너쳐 기반 침입탐지 기술은 과탐지(false positive)가 많고 새로운 공격이나 변형된 유형의 공격을 감지하기 어렵다 우리는 앞선 논문[1]을 통해 시그너쳐 기반 침입 탐지 시스템과 기계학습 기반 침입 탐지 시스템을 Alpha-cut 방법을 이용하여 결합한 모델을 제시 하였다. 본 논문은 Alpha-cut의 후속연구로 기존 모델에서 감지하지 못하는 미탐지(false negative)를 줄이기 위한 Beta-pick 방법을 제안한다. Alpha-cut은 시그너쳐 기반 침입탐지 시스템의 공격 탐지결과에 대한 정확성을 높이는 방법인 반면에, Beta-rick은 공격을 정상으로 판단하는 경우를 줄이는 방법이다. Alpha-cut과 Beta-pick을 위해 사용된 기계학습 알고리즘은 XIBL(Extended Instance based Learner)이며, C4.5를 적용했을 때와 차이점을 결과로서 제시한다. 제안한 방법의 효과를 설명하기 위해 시그너쳐 기반 침입탐지 시스템의 탐지결과에 Alpha-cut과 Beta-pick을 적용하여 오경보(false alarm)가 감소함을 보였다.

온사이트 지진조기경보를 위한 딥러닝 기반 실시간 오탐지 제거 (Deep Learning-Based, Real-Time, False-Pick Filter for an Onsite Earthquake Early Warning (EEW) System)

  • 서정범;이진구;이우동;이석태;이호준;전인찬;박남률
    • 한국지진공학회논문집
    • /
    • 제25권2호
    • /
    • pp.71-81
    • /
    • 2021
  • This paper presents a real-time, false-pick filter based on deep learning to reduce false alarms of an onsite Earthquake Early Warning (EEW) system. Most onsite EEW systems use P-wave to predict S-wave. Therefore, it is essential to properly distinguish P-waves from noises or other seismic phases to avoid false alarms. To reduce false-picks causing false alarms, this study made the EEWNet Part 1 'False-Pick Filter' model based on Convolutional Neural Network (CNN). Specifically, it modified the Pick_FP (Lomax et al.) to generate input data such as the amplitude, velocity, and displacement of three components from 2 seconds ahead and 2 seconds after the P-wave arrival following one-second time steps. This model extracts log-mel power spectrum features from this input data, then classifies P-waves and others using these features. The dataset consisted of 3,189,583 samples: 81,394 samples from event data (727 events in the Korean Peninsula, 103 teleseismic events, and 1,734 events in Taiwan) and 3,108,189 samples from continuous data (recorded by seismic stations in South Korea for 27 months from 2018 to 2020). This model was trained with 1,826,357 samples through balancing, then tested on continuous data samples of the year 2019, filtering more than 99% of strong false-picks that could trigger false alarms. This model was developed as a module for USGS Earthworm and is written in C language to operate with minimal computing resources.

카메라 기반 객체의 위치인식을 위한 왜곡제거 및 오검출 필터링 기법 (Distortion Removal and False Positive Filtering for Camera-based Object Position Estimation)

  • 진실;송지민;최지호;진용식;정재진;이상준
    • 대한임베디드공학회논문지
    • /
    • 제19권1호
    • /
    • pp.1-8
    • /
    • 2024
  • Robotic arms have been widely utilized in various labor-intensive industries such as manufacturing, agriculture, and food services, contributing to increasing productivity. In the development of industrial robotic arms, camera sensors have many advantages due to their cost-effectiveness and small sizes. However, estimating object positions is a challenging problem, and it critically affects to the robustness of object manipulation functions. This paper proposes a method for estimating the 3D positions of objects, and it is applied to a pick-and-place task. A deep learning model is utilized to detect 2D bounding boxes in the image plane, and the pinhole camera model is employed to compute the object positions. To improve the robustness of measuring the 3D positions of objects, we analyze the effect of lens distortion and introduce a false positive filtering process. Experiments were conducted on a real-world scenario for moving medicine bottles by using a camera-based manipulator. Experimental results demonstrated that the distortion removal and false positive filtering are effective to improve the position estimation precision and the manipulation success rate.

구순문 인식을 위한 복수 해상도 시스템의 패턴 커널에 관한 연구 (On Pattern Kernel with Multi-Resolution Architecture for a Lip Print Recognition)

  • 김진옥;황대준;백경석;정진현
    • 한국통신학회논문지
    • /
    • 제26권12A호
    • /
    • pp.2067-2073
    • /
    • 2001
  • 생체인식 시스템은 인간 고유의 물리적인 특성들을 이용하여 인간을 인식하는 기술이다. 그리고 특정한 물리적인 특징들을 얻기 위하여 센서를 사용하고 디지털 패턴으로 변환시켜서 저장된 패턴과 비교한다. 구순문 인식은 지문, 음싶 패턴, 홍태 패턴과 얼굴 인식과 같은 신체적 특징에 비하여 상대적으로 연구가 많이 이루어지지 않았다. 구순문은 CCD 카메라를 이용할 경우 홍채나 얼굴 패턴 같은 다른 특징 요소와 연결하여 인식 시스템을 구축할 수 있는 장점을 가지고 있다. 구순문 인식을 위해 패턴 커널을 이용한 새로운 방법을 제시하였다. 패턴 커널은 여러 개의 국부 구순문 마스크 (local lip print mask)들로 구성된 함수이며, 구순문의 정보를 디지털 데이터로 전환시켜 준다. 복수 해상도를 가지는 인식 시스템은 단일 해상도의 시스템보다 더욱 신뢰적 이며 인식률도 높다. 복수해상도 구조는 오인식률을 현저히 감소시키므로 복수해상도를 갖는 구순문 인식은 생체인식 시스템에 잘 활용될 수 있다.

  • PDF

2005년 동아시아 지역에서 발생한 모래폭풍과 먼지침전(황사)의 관측 (On the Observation of Sandstorms and Associated Episodes of Airborne Dustfalls in the East Asian Region in 2005)

  • 김학성;정용승
    • 한국지구과학회지
    • /
    • 제30권2호
    • /
    • pp.196-209
    • /
    • 2009
  • 2005년 1월부터 12월까지 지속적으로 몽골과 중국 북부의 사막과 건조지대에서 모래폭풍의 발생, 이동 그리고 한국에서 먼지침전(황사)을 모니터링하였다. NOAA 위성의 AVHRR(Advanced Very High Resolution Radiometer) 관측 자료를 직접 수신하여 false colour 합성영상을 만들어 모래폭풍의 발생과 이동 사례를 중점 분석하였다. 또한 모래폭풍 발원지의 풍하측에 위치한 한국 중부의 청원에서 PM10 농도가 $190{\mu}g\;m^{-3}$ 이상인 황사 사례에 대한 PM10, PM2.5농도와 시정 그리고 $O_3,\;NO_2,\;CO,\;SO_2$ 농도의 변동을 분석하였다. 2005년에는 발원지에 적설과 강력한 고기압, 저기압의 이동이 미흡하여 강한 모래폭풍의 발생이 적었고 한국에서도 황사의 발생 빈도가 1997-2005년의 평균과 비교하여 적었다. 한국에는 2005년 총 7회, 11일간의 황사 사례가 발생하였다. 여름에는 발원지가 습하고 바람이 약해 모래폭풍의 발생이 적었으며, 한국에서도 황사의 발생이 없었다. 발원지에서 발생한 모래폭풍이 중국의 대도시와 산업지대를 거치지 않고 한국으로 직접 이동할 경우, 먼지침전에 따른 PM2.5 농도는 PM10 농도의 20% 이하였다. 그러나 모래폭풍이 중국 동부의 산업지역을 거쳐 한국으로 이동할 경우 인위적 대기오염 물질이 함께 유입되어 PM2.5 농도가 증가하여 PM10농도의 25%이상을 나타내었다. 또한 2005년에는 사막에서 날아온 모래 먼지의 PM10농도가 $190{\mu}g\;m^{-3}$ 이하여서 황사로는 미흡한 5개 이상 사례를 관측 및 분석하였다.

Ideology, Politics, and Social Science Scholarship on the Responsibility of Intellectuals

  • Koerner, E.F.K.
    • 인문언어
    • /
    • 제2권2호
    • /
    • pp.51-84
    • /
    • 2002
  • The 1990s have seen the publication of many books devoted to Language and Ideology (cf. Joseph & Taylor 1990. for one of the early ones) even though the term 'ideology' itself has remained ill-defined (Woolard 1998). The focus of attention has usually been placed on the particular use of language and often for some kind of 'political' ends, not on linguistic or other scholarship which might have been driven by some sort of ideology, i.e., a bundle of assumptions which themselves were taken as given. At least since Edward Said's 1978 book Orientalism, it has been clear to everyone that scholars construct their conceptualization of things in line with their understanding of the cultural, social, and political world in which they live, and that this often unreflected 'pre-understanding' effects their view of cultures that are different from theirs and more often than not geographically and temporally distant from theirs. This recognition has had a sobering effect no doubt, and Said's book has long since become 'mainstream.' Much more disturbing to the scholarly profession has been the publication of Martin Bernal's Black Athena in 1987, since it went much further, going beyond accusations of colonialism and cultural bias, in suggesting that the Western representation of Classical Greece over the past two hundred years was false and that what had been accepted until now about occidental antiquity must now be seen derived from African-Asiatic cultures of the Near East, notably that of the Ancient Egyptians, and that no other than Socrates should be seen as black man. While we may understand the intellectual climate in the United States that led academics to present 'myth as history' (Lefkowitz 1996), it is obvious that lines of regular scholarly principles of investigation have been crossed (cf Lefkowitz & Rogers 1996). The present paper investigates what may be seen as the ideological underpinnings of such work. After reviewing some recent scholarship in the area of linguistic historiography that have shown that academic work has never been 'value-neutral' (as may have been assumed or has been claimed by some practitioners), it is argued that in effect one must be aware of what Clemens Knobloch has recently termed Resonanzbedarf, i.e., the desire, whether conscious or not, of scholars-and probably scientists, too-to have their work recognized by the educated public and that, in so doing, their discourses tend to pick up on contemporary popular notions. These efforts may be harmless if everyone was to recognize these allusions and adoption of certain lexical. items(buzz words) as props or what Germans call Versatzstiicke, but history tells us that this has not always been the case. Still, as Hutton (1999) has shown, not all scholarship during the Third Reich for example can simply be dismissed as worthless because it was conducted in under a prevailing political ideology. Indeed, in seemingly innocent times, linguists can be shown to frame their argument in a way that makes them appear so utterly superior to their predecessors (cf. Lawson 2001). Upon closer inspection, those discourses turn out to be much like those of scholars in nationalistic environments that have tended to select their 'facts' to prove a particular hypothesis (cf., e.g., Koerner 2001). The article argues for scholars to take a more active role in exploding myths, scientifically unfounded claims, and ideologically driven distortions, especially those that are socially and politically harmful.

  • PDF