• 대한인간공학회:학술대회논문집
• /
• 대한인간공학회 1993년도 추계학술대회논문집
• /
• pp.124-132
• /
• 1993

This study compared two prediction methods-regression and artificial neural network (ANN) on the visual search performance when monitoring a multi-parameter screen with different occurrence frequencies. Under the highlighting condition for the highest occurrence frequency parameter as a search cue, it was found from the requression analysis that variations of mean search time (MST) could be expained almost by three factors such as the number of parameters, the target occurrence frequency of a highlighted parameter, and the highlighted parameter size. In this study, prediction performance of ANN was evaluated as an alternative to regression method. Backpropagation method which was commonly used as a pattern associator was employed to learn a search behavior of subjects. For the case of increased number of parameters and incresed target occurrence frequency of a highlighted parameter, ANN predicted MST's moreaccurately than the regression method (p<0.000). Only the MST's predicted by ANN did not statistically differ from the true MST's. For the case of increased highlighted parameter size. both methods failed to predict MST's accurately, but the differences from the true MST were smaller when predicted by ANN than by regression model (p=0.0005). This study shows that ANN is a good predictor of a visual search performance and can substitute the regression method under certain circumstances.

• 한국농공학회논문집
• /
• 제60권1호
• /
• pp.47-57
• /
• 2018

This study implemented the rainfall-runoff analysis of the Mekong River basin using the SWAT (Soil and Water Assessment Tool). The runoff analysis was simulated for 2000~2007, and 11 parameters were calibrated using the SUFI-2 (Sequential Uncertainty Fitting-version 2) algorithm of SWAT-CUP (Calibration and Uncertainty Program). As a result of analyzing optimal parameters and sensitivity analysis for 6 cases, the parameter ALPHA_BF was found to be the most sensitive. The reproducibility of the rainfall-runoff results decreased with increasing number of stations used for parameter calibration. The rainfall-runoff simulation results of Case 6 showed that the RMSE of Nong Khai and Kratie stations were 0.97 and 0.9, respectively, and the runoff patterns were relatively accurately simulated. The runoff patterns of Mukdahan and Khong Chaim stations were underestimated during the flood season from 2004 to 2005 but it was acceptable in terms of the overall runoff pattern. These results suggest that the combination of SWAT and SWAT-CUP models is applicable to very large watersheds such as the Mekong for rainfall-runoff simulation, but further studies are needed to reduce the range of modeling uncertainty.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 추계학술대회
• /
• pp.441-443
• /
• 2018

본 논문은 RGB 영상 이용하여 하지 움직임에 대한 분석을 다룬다. 딥러닝 접근방법인 객체 인식 Segmentation 알고리즘과 자세 검출 알고리즘을 융합한 방법과 BMC(Background Model Challenge)을 활용하여 RGB 영상을 보행 분석 요소로 사용하였다. 본 연구에서 제시한 영상 보행 분석은 보행패턴 인식과 비정상적인 보행 등의 분류를 위한 변수로서 활용할 수 있을 것으로 판단된다.

• 한국콘텐츠학회논문지
• /
• 제10권4호
• /
• pp.10-18
• /
• 2010

본 논문에서는 유전자알고리즘을 이용한 시그모이드 활성화 함수 파라미터의 최적화와 이중나선기준문제(two spirals benchmark problem)의 입력공간 패턴인식 상태를 분석 한다. 실험을 위하여 캐스케이드 코릴레이션 학습 알고리즘(Cascade Correlation learning algorithm)을 이용한다. 첫 번째 실험에서는 기본적인 시그모이드 활성화 함수를 사용하여 이중나선 문제를 분석하고, 두 번째 실험에서는 시그모이드 활성화 함수(sigmoidal activation function)의 파라미터 값이 서로 다른 함수를 사용하여 8개의 풀을 구성한다. 세 번째 실험에서는 시그모이드 함수의 변위를 결정하는 세 개의 파라미터 값을 유전자 알고리즘을 이용하여 얻고 이 파라미터 값들이 적용된 시그모이드 함수들은 후보뉴런의 활성화를 위해서 사용된다. 이러한 알고리즘의 성능평가를 위하여 각 학습단계 마다 입력패턴공간에서 인식된 이중나선의 형태를 보여준다.

• 센서학회지
• /
• 제13권6호
• /
• pp.436-445
• /
• 2004

In this study, a uroflowmetry system was developed to detect a voiding symptom conveniently at home or hospital. A implemented hardware was composed of mechanism and system circuit part, the software was developed to process uroflow data, graph display, extraction of parameter, and evaluation of congregate rate so as to analysis obtaining uroflow data. The following experiment was performed to evaluate an ability of classification and fitness. The curve pattern of uroflow was classified into each symptom. Various parameters were calculated in the curve pattern of each uroflow as follows. The parameters are MFR, AFR, VOL, VT, and FT. A significant difference among parameters was examined by a statistical analysis for extracted parameters between normal and abnormal experimental group. The uroflow data with the various symptom was divided into normal and abnormal group using fuzzy classifier. The result of the fuzzy classification using MFR and AFR was superior by 91.23 % than grouping evaluation including VOL.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.115-115
• /
• 2000

A gait pattern classification method using electromyography(EMG) signal is presented. The gait pattern with four stages such as stance, heel-off, swing and heel-strike is analyzed and classified using feature parameters such as zero-crossing, integral absolute value and variance of the EMG signal. The EMG signal from Tibialis Anterior and Gastrocnemius muscles was obtained using the surface electrodes, and low-pass filtered at 10kHz. The filtered analog signal was sampled at every 0.5msec and converted to digital signal with 12-bit resolution. The obtained data is analyzed and classified in terms of feature parameters. Analysis results are given to show that the gait patterns classified by the proposed method are feasible.

• 한국음향학회지
• /
• 제26권5호
• /
• pp.207-213
• /
• 2007

본 연구에서는 스펙트럼 분석과 신경망을 이용한 효과적인 음성/음악 분류 방법을 제안한다. 제안하는 방법은 스펙트럼을 분석하여 스펙트럴 피크 트랙에서 지속성 특징 파라미터인 MSDF(Maximum Spectral Duration Feature)를 추출하고 기존의 특징 파라미터인 MFSC(Mel Frequency Spectral Coefficients)와 결합하여 음성/음악 분류기의 특징으로 사용한다. 그리고 신경망을 음성/음악 분류기로 사용하였으며, 제안하는 방법의 성능 평가를 위해 학습 패턴 선별과 양, 신경망 구성에 따른 다양한 성능 평가를 수행하였다. 음성/음악 분류 결과 기존의 방법에 비해 성능 향상과 학습 패턴의 선별과 모델 구성에 따른 안정성을 확인할 수 있었다. MSDF와 MFSC를 특징 파라미터로 사용하고 50초 이상의 학습 패턴을 사용할 때 음성에 대해서는 94.97%, 음악에 대해서는 92.38%의 분류율을 얻었으며, MFSC만 사용할 때보다 음성은 1.25%, 음악은 1.69%의 향상된 성능을 얻었다.

• 대한기계학회논문집B
• /
• 제26권4호
• /
• pp.622-628
• /
• 2002

Numerical analysis has been carried out for two-dimensional steady and unsteady mixed convection in the annulus between co-rotating horizontal cylinders with a heated inner cylinder. The ratio of annulus gap($\sigma$) is taken from 1 to 10 and the order of mixed-convection parameter B(=Gr/(1+Re)$^2$) varies from 10$^4$to $10^0$. The flow patterns over this parameter range are steady multicellular, oscillatory multicellular or steady unicellular. The addition of co-rotating of both cylinders stabilizes the flow in the annulus and weakens the unsteadiness. Even in the large values of rotating parameter such as of $10^0$/($\sigma$=2) and 10$^2$($\sigma$=10), the flow pattern becomes asymptotic to the steady unicellular flow, like as in the rigid-body rotating flow.

• 한국환경보건학회지
• /
• 제33권4호
• /
• pp.299-305
• /
• 2007

Distribution of fecal pollution indicator bacteria and environmental parameter were investigated of urban artificial lakes. An average concentration of temperature, pH, SS, DO, $COD_{Mn}$, T-P, T-N, Turbidity, Chl-a were $21.5^{\circ}C$, 8.07, 116.70 mg/l, 8.66 mg/l, 2.24 mg/1, 0.52 mg/l, 1.71mg/l, 80.54 NTU, and 52.12 mg/l respectively. From the results of bivariate correlation analysis, fecal contamination indicator bacteria were found to be mutually correlated. And turbidity and suspended solid were correlated. From the results of principal component analysis, four factors were extracted. And four factors of variance explained up to 81.5 percentage. Factor 1 was pollution pattern by fecal contamination, factor 2 was physical pollution pattern by pollution source, factor 3 was natural pollution by precipitation, and factor 4 was artificial pollution pattern by organism.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2005년도 추계학술대회 논문집
• /
• pp.593-597
• /
• 2005

FIB equipment can perform sputtering and chemical vapor deposition simultaneously. It is very advantageously used to fabricate a micro structure part having 3D shape because the minimum beam size of ${\phi}$ 10nm and smaller is available. Currently FIB is not being applied in the fabrication of this micro part because of some problems to redeposition and charging effect of the substrate causing reduction of accuracy with regards to shape and productivity. Furthermore, the prediction of the material removal rate information should be required but it has been insufficient for micro part fabrication. The paper have the targets that are FIB-CVD characteristic analysis and minimum line pattern resolution achievement fur 3D micro fabrication. We make conclusions with the analysis of the results of the experiment according to beam current, pattern size and scanning parameters. CVD of 8 pico ampere shows superior CVD yield but CVD of 1318 pico ampere shows the pattern sputtered. And dwell time is dominant parameter relating to CVD yield.