• Title/Summary/Keyword: Flow Detection

Search Result 1,247, Processing Time 0.039 seconds

Abnormal Behavior Detection and Localization Using Aspect Ratio Based on Mask R-CNN (Mask R-CNN 기반 Aspect Ratio를 활용한 이상행동 검출 및 영역화 방법)

  • Lim, Hyunseok;Hu, Xufeng;Gwak, Jeonghwan
    • 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.

  • PDF

Physics-Based Cloth and Liquid Interaction using GPU Optimization (GPU 최적화를 이용한 물리 기반 옷감과 액체의 상호작용)

  • Seong-Hyeok Moon;Jong-Hyun Kim
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2023.01a
    • /
    • pp.395-398
    • /
    • 2023
  • 본 논문에서는 물리 기반 옷감 시뮬레이션과 SPH(Smoothed particle hydrodynamics) 기반의 유체 시뮬레이션 간의 상호작용에서 표현되는 다양한 물리적 효과를 GPU 기반으로 빠르게 표현할 수 있는 프레임워크를 제안한다. 기존 기법과는 다르게 수치적 안정성을 개선하기 위해 CCD(Continuous collision detection)를 활용하였으며, 모든 연산이 GPU에서 동작하기 때문에 매우 빠르게 옷감과 유체의 상호작용 장면인 다공성 재질, 기공 흐름, 흡수, 방사, 확산을 모델링할 수 있다.

  • PDF

A Study on Anomaly Detection Neural Network Model Based On Flow Direction/Velocity Data (유향/유속 데이터 중심의 이상 검출 신경망 모델)

  • Seong-Kil Hyun;Dong-Young Yoo
    • Annual Conference of KIPS
    • /
    • 2023.05a
    • /
    • pp.555-557
    • /
    • 2023
  • 해양의 영향을 많이 받는 우리나라의 지리적 특성상 해양 상황은 산업 및 생업과 밀접한 관계가 있다. OPEN API 를 이용하여 유향/유속, 조위등 해양 환경관련 실시간 시계열 데이터를 수집한 후 2 차원 공간에 표시하여 순환 신경망 모델을 이용하여 학습한다. 학습된 모델을 실제 데이터에 적용하여 파랑을 예측한다. 시계열의 성격이 있고 공간상에 표시할 수 있는 데이터라면 본 논문에서 제시한 체계를 통해 예측할 수 있을 것이라 예상한다.

Smart Contract Vulnerability Detection Study Based on Control Flow Graphs (제어 흐름 그래프 기반 스마트 컨트랙트 취약성 탐지 연구)

  • Yoo-Young Cheong;La Yeon Choi;Dong-Hyuk Im
    • Annual Conference of KIPS
    • /
    • 2023.11a
    • /
    • pp.1247-1249
    • /
    • 2023
  • 스마트 컨트랙트는 블록체인 상에서 실행되는 프로그램으로 복잡한 비즈니스 논리를 처리할 수 있다. 그러나 블록체인의 무결성과 조건에 따라 실행되는 특성을 이용한 악의적 사용으로 인하여 블록체인 보안에서 시급한 문제가 되고있다. 따라서 스마트 컨트랙트 취약성 탐지문제는 최근 많은 연구가 이루어지고 있다. 그러나 기존 연구의 대부분이 단일 유형의 취약성 여부에 대한 탐지에만 초점이 맞춰져 있어 여러 유형의 취약성에 대한 동시 식별이 어렵다. 이 문제를 해결하고자 본 연구에서는 스마트 컨트랙트 소스코드 제어 흐름 그래프를 기반으로 그래프의 forward edge와 backward edge를 고려한 신경망으로 그래프 구조를 학습한 후 그래프 multi-label classification을 진행하여 다중 취약성을 탐지할 수 있는 모델을 제안한다.

A Method for Real-Time Face Detection through Optical Flow and Scale Resampling (광학 흐름과 스케일 리샘플링을 통한 실시간 얼굴 탐지 기법)

  • Sang-Jeong Kim;Dong-Gun Lee;Yeong-Seok Seo
    • Annual Conference of KIPS
    • /
    • 2024.05a
    • /
    • pp.862-863
    • /
    • 2024
  • 기존의 딥러닝 모델을 활용한 얼굴 탐지 시스템은 영상을 처리할 때 이미지의 양이 과도하여 추론 속도가 영상 재생 속도보다 느려지게 되고, 이로 인해 지연 현상이 발생한다. 본 논문은 이미지 크기 조정 및 광학 흐름을 활용하여 얼굴 탐지에 필요한 추론량을 줄이는 기법을 제안한다. 제안된 기법은 세 단계의 처리 과정으로 구성된다. 첫 번째 단계에서는 프레임의 크기를 줄여 프레임 처리 속도를 효과적으로 향상시킨다. 두 번째 단계에서는 비탐지 구간이 아닌 프레임만을 배치 처리하여 딥러닝 모델로 추론하여 처리 시간을 단축시킨다. 세 번째 단계에서는 광학 흐름 알고리즘을 이용하여 비탐지 구간에서 얼굴 추적을 함으로써 정확도는 유지하면서 탐지 시간을 단축한다. 본 논문에서 제안하는 이미지 크기 조정 및 광학 흐름 알고리즘 기반 얼굴 탐지 시스템은 처리 시간을 수십 배 이상 단축하여 영상에서의 얼굴 탐지에 있어서 우수한 성능을 입증하였다.

HPLC Method for Simultaneous Quantitative Detection of Quercetin and Curcuminoids in Traditional Chinese Medicines

  • Ang, Lee Fung;Yam, Mun Fei;Fung, Yvonne Tan Tze;Kiang, Peh Kok;Darwin, Yusrida
    • Journal of Pharmacopuncture
    • /
    • v.17 no.4
    • /
    • pp.36-49
    • /
    • 2014
  • Objectives: Quercetin and curcuminoids are important bioactive compounds found in many herbs. Previously reported high performance liquid chromatography ultraviolet (HPLC-UV) methods for the detection of quercetin and curcuminoids have several disadvantages, including unsatisfactory separation times and lack of validation according the standard guidelines of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. Methods: A rapid, specific, reversed phase, HPLC-UV method with an isocratic elution of acetonitrile and 2% v/v acetic acid (40% : 60% v/v) (pH 2.6) at a flow rate of 1.3 mL/minutes, a column temperature of $35^{\circ}C$, and ultraviolet (UV) detection at 370 nm was developed. The method was validated and applied to the quantification of different types of market available Chinese medicine extracts, pills and tablets. Results: The method allowed simultaneous determination of quercetin, bisdemethoxycurcumin, demethoxycurcumin and curcumin in the concentration ranges of $0.00488-200{\mu}g/mL$, $0.625-320{\mu}g/mL$, $0.07813-320{\mu}g/mL$ and $0.03906-320{\mu}g/mL$, respectively. The limits of detection and quantification, respectively, were 0.00488 and $0.03906{\mu}g/mL$ for quercetin, 0.62500 and $2.50000{\mu}g/mL$ for bisdemethoxycurcumin, 0.07813 and $0.31250{\mu}g/mL$ for demethoxycurcumin, and 0.03906 and $0.07813{\mu}g/mL$ for curcumin. The percent relative intra day standard deviation (% RSD) values were $0.432-0.806{\mu}g/mL$, $0.576-0.723{\mu}g/mL$, $0.635-0.752{\mu}g/mL$ and $0.655-0.732{\mu}g/mL$ for quercetin, bisdemethoxycurcumin, demethoxycurcumin and curcumin, respectively, and those for intra day precision were $0.323-0.968{\mu}g/mL$, $0.805-0.854{\mu}g/mL$, $0.078-0.844{\mu}g/mL$ and $0.275-0.829{\mu}g/mL$, respectively. The intra day accuracies were 99.589%-100.821%, 98.588%-101.084%, 9.289%-100.88%, and 98.292%-101.022% for quercetin, bisdemethoxycurcumin, demethoxycurcumin and curcumin, respectively, and the inter day accuracy were 99.665%-103.06%, 97.669%-103.513%, 99.569%-103.617%, and 97.929%-103.606%, respectively. Conclusion: The method was found to be simple, accurate and precise and is recommended for routine quality control analysis of commercial Chinese medicine products containing the flour flavonoids as their principle components in the extracts.

Direction of Development of Reaction to Bio-terrorism (생물테러리즘 대응을 위한 기술적 측면의 발전방향)

  • Lee, Kwang-Iyeol;Kim, Chang-Ho
    • Korean Security Journal
    • /
    • no.14
    • /
    • pp.311-336
    • /
    • 2007
  • This study examines the concepts of counter-act against the Bio-terrorism and the reaction system in advanced countries, thereby to find out reaction system necessary to Korea. Acts on anti-Bio-terrorism is divided to detection stage, protection stage, diagnosis stage and detoxication and neutralization stage according to flow of event occurrence. As for detection stage, Korea is developing it as contact type, while advanced countries are under development of the devices that may detect the terrorism from the remote distance. It is necessary for Korea to develop the remote-distant detection system as well as the contact type of device that may promptly operate. Among the protection gears, the quality of Korea's gas mask is recognized worldwide, but that of other outfits should be improved by applying the state-of-art science technology. The diagnosis device also should be developed to the extent that the dispatched initial action team may make immediate decisions necessary in the field. As the current trends for detoxication materials worldwide require the improvement to new materials harmless to human body and equipment, Korea is also required to acquire those materials. The technology for neutralization means the development of vaccine and antibiotics and it requires the development made by shared efforts worldwide. For this purpose, it is necessary to further develop Korea's medical technology. In addition, the further efforts are required in terms of reaction manual, training model, public communication efforts and preparation for trauma syndrome.

  • PDF

Optimal HPLC Condition for Simultaneous Determination of Anthocyanins in Black Soybean Seed Coats (검정콩 함유 안토시아닌의 동시분석을 위한 최적 HPLC 분석 조건)

  • Choung, Myoung-Gun
    • KOREAN JOURNAL OF CROP SCIENCE
    • /
    • v.53 no.4
    • /
    • pp.359-368
    • /
    • 2008
  • Black soybean has been widely utilized as foods and oriental medicinal materials. The pigmentation in the seed coat of black soybean is due to accumulate anthocyanins in the epidermis palisade layer. The anthocyanins of black soybean seed coat are considered as a parameter of quality evaluation of black soybean. Therefore, the purpose of this study was to investigate the most suitable HPLC condition for simultaneous determination of anthocyanins in black soybean seed coats extracts. The efficient HPLC analytical condition of D3G, C3G, and Pt3G contained extracts of black soybean seed coats was developed. The gradient elution employed a $250\;mm\;{\times}\;4.6\;mm$ i.d. YMC-pak ODS-AM 303 column. The gradient system was used two mobile phases. A gradient elution was performed with mobile phase A, consisting of 5% aqueous formic acid, and mobile phase B, comprising 5% formic acid - acetonitrile, and delivered at a flow rate of 0.7 mL/min as follows: $0{\sim}35\;min$, $90%\;A{\sim}60%\;A$; 36 min, 90% A; 46 min, 90% A. The UV-VIS. detection wavelength was set at 520 nm. The limit of detection (LOD) for D3G, C3G, and Pt3G were under 10 ng/mL.

A Water Surface Detection Method by Correlation Analysis of Watermark Images with Time Interval (시차가 있는 수위표 이미지의 상관분석을 통한 수면측정기법)

  • Seo, Myoung-Bae;Lee, Chan-Joo;Kim, Dong-Gu
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.14 no.1
    • /
    • pp.470-477
    • /
    • 2013
  • The aim of this study is to suggest a detection method of water surface location and its evaluation results of application for same vertical position in two successive images with time interval including both staff gauge and water surface. A specific rectangular inspection area is defined from the top of watermark and then the correlation coefficients for the inspection area of the same position of two images with short time interval is calculated. Accordingly, it is possible to identify differences between changing area and fixed area of pixel density by the water flow. The photographs taken in the laboratory were analyzed in order to validate the proposed technique. As the result of the experiment, it is identified that characteristic of correlation coefficients depends on the size of the inspection area. In the case that the inspection area is within the entire width of the watermark, water surface characteristic according to correlation coefficients is clearly noticeable. Thus, it is identified that the proposed technique can be utilized to search water surfaces. Besides, using corelation analysis of two images with time interval, it is identified that error range between 10 and 42cm was reduced in the level of 2.6cm or less in the contaminated photo of existing image stage gauge. Therefore, it is expected that the suggested method can be utilized to enhance image stage gauge performance improving the previous water surface detection method.

Validation of an analytical method of dieckol for standardization of Ecklonia cava extract as a functional ingredient (감태추출물의 기능성원료 표준화를 위한 지표성분 dieckol의 분석법 검증)

  • Xu, Yan;Kim, Eun Suh;Lee, Ji-Soo;Kim, Gun-Hee;Lee, Hyeon Gyu
    • Korean Journal of Food Science and Technology
    • /
    • v.51 no.5
    • /
    • pp.420-424
    • /
    • 2019
  • An HPLC analysis method was developed and standardized for the detection of dieckol as a functional food ingredient in Ecklonia cava extracts. HPLC was performed using a Capcell Pak C18 column ($250{\times}4.6mm$, $5{\mu}m$) with a gradient elution of water and acetonitrile, both containing 0.1% (v/v) trifluoroacetic acid, at a flow rate of 1.0 mL/min at $25^{\circ}C$. The eluate was detected at 230 nm. For validation, the specificity, linearity, accuracy, precision, limit of detection (LOD), and limit of quantification (LOQ) of dieckol were measured. The calibration curve for the detection of dieckol had high linearity ($R^2=0.9994$), with LOD and LOQ values of 0.38 and $1.16{\mu}g/mL$, respectively. Recovery of the quantified compound ranged from 99.61 to 100.71%. The relative standard deviation values of the intra-day and inter-day precisions were less than 1.7 and 1.25%, respectively. These results indicate that the reported HPLC method is simple, reliable, and reproducible for the detection of dieckol in Ecklonia cava extracts.