• 산업기술연구
• /
• 제21권B호
• /
• pp.59-66
• /
• 2001

In this paper, a novel overmodulation strategy for PWM inverters to extend linearity region is presented. The proposed strategy uses the concept of space-vector PWM(SVPWM) based on the zero sequence signal(offset voltage) injection principle. So, by modifying the pole voltage simply, the linear control of inverter output voltage over the whole overmodulation range can be achieved easily. The proposed strategy is so simple that its practical implementation is easy. The validity of the proposed strategy is confirmed by the experimental results.

• 대한산업공학회지
• /
• 제23권4호
• /
• pp.825-833
• /
• 1997

The purpose of this paper is to develop a method of postoptimality analysis that can be applied to an optimal solution of a maximum flow problem. We first use the transformed network corresponding to a given network. In such a network we conduct postoptimality analysis by determining changes in the optimal solution precipitated by changes in the capacity as the arc capacity varies from 0 to infinite. By this method we can easily calculate not only the characteristic region where the given optimal solution remains unchanged, but also the characteristic region where the value of the maximal flow gradually increases or decreases. The proposed method is demonstrated by numerical example.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.26-29
• /
• 2005

In order to compute cloud coverage statistics over Asian region, an operational scheme for masking cloud-contaminated pixels in Advanced Very High Resolution Radiometer (AVHRR) daytime data was developed, evaluated and presented. Dynamic thresholding was used with channell, 2 and 3 to automatically create a cloud mask for a single image. Then the IO-day cloud coverage imagery was generated over the whole Asian region along with cloud-free composite imagery. Finally the monthly based statistics were computed based on the derived cloud coverage imagery in terms of land cover and country. As a result, it was found that 20-day is required to acquire the cloud free data over the whole Asia using NOAA AVHRR. The to-day cloud coverage and cloud-free composite imagery derived in this research is available via the web-site http://webpanda.iis.u-tokyo.ac.jp/CloudCover/.

• 전기학회논문지
• /
• 제67권7호
• /
• pp.865-870
• /
• 2018

A machine vision based industrial inspection includes defects detection and classification. Fast inspection is a fundamental problem for many applications of real-time vision systems. It requires little computation time and localizing defects robustly with high accuracy. Deep learning technique have been known not to be suitable for real-time applications. Recently a couple of fast region-based CNN algorithms for object detection are introduced, such as Faster R-CNN, and YOLOv2. We apply these methods for an industrial inspection problem. Three CNN based detection algorithms, VOV based CNN, Faster R-CNN, and YOLOv2, are experimented for defect detection on metal surface. The results for inspection time and various performance indices are compared and analysed.

• 한국산학기술학회논문지
• /
• 제7권4호
• /
• pp.575-579
• /
• 2006

본 논문에서는 전동기 시스템의 극저속 영역에서 안정적이고 동특성이 우수한 운전을 하기 위해 전차원상태 관측기로부터 추정되는 속도정보를 입력으로 하는 최소차원 토크 관성관측기를 구성하여 부하외란과 관성을 피드포워드 제어를 함으로써 극저속영역에 대한 강인성을 갖도록 한다. 제시한 제어 알고리즘을 MATLAB SIMULINK를 통해 전체 시스템을 모델링하고 실험을 통해 극저속에서 속도 추정 능력이 향상됨을 확인 한다.

• 산업기술연구
• /
• 제28권B호
• /
• pp.143-148
• /
• 2008

A real-time face detection is to find human faces robustly under the cluttered background free from the effect of occlusion by other objects or various lightening conditions. We propose a face detection system for real-time applications using wavelet decomposition method based on Gabor features. Firstly, skin candidate regions are extracted from the given image by skin color filtering and projection method. Then Gabor-feature based template matching is performed to choose face cadidate from the skin candidate regions. The chosen face candidate region is transformed into 2-level wavelet decomposition images, from which feature vectors are extracted for classification. Based on the extracted feature vectors, the face candidate region is finally classified into either face or nonface class by the Levenberg-Marguardt back-propagation neural network.

• 산업기술연구
• /
• 제13권
• /
• pp.25-31
• /
• 1993

This paper presents an analysis of microwave device component. An H-Plane waveguide component with arbitrary shape is analyzed using finite element method(FEM) cooperated with boundary element method(BEM). The finite element method(FEM) is applied to the junction region and the boundary element method(BEM) to the waveguide region. For the application of BEM in the waveguide structure, a ray representation of the waveguide Green's function is used. The proposed technique was applied to the analysis of the waveguide inductive junction to compare the numerical result with the result of the mode matching technique. The comparison showed good agreements between the two results. Transmitted powers were also computed in T-junction waveguides for the various shape of the junction area.

• 산업기술연구
• /
• 제30권B호
• /
• pp.91-98
• /
• 2010

Recently, there has been enormous damage due to river floodings caused by localized heavy rains. The direct discharge triggered by those torrential rains inflicts severe property damage on the residents of nearby areas. To minimize the possibility of river floodings in case of heavy rains and to predict the possible damage, the management of existing rainfall and water level observatories should be checked and prediction methods based on the characteristics of water usage and floodgate of nearby rivers must be further analyzed. Therefore, this research analyzed the water level change predictions on different spots with a regression equation of rainfall and water levels, using the observation data of the water level observatory in Jeongseon-gun, Gangwon Province and the rainfall observatory which are located on the upper region of the Han river.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제17권1호
• /
• pp.47-54
• /
• 2013

This work deals with automatic motion detection for with surveillance tracking that aims to provide high-lighting movable objects which is discriminated from moving backgrounds such as moving trees, etc. For this aim, we perform a false background region detection together with an initial foreground detection. The false background detection detects the moving backgrounds, which become eliminated from the initial foreground detection. This false background detection is done by performing the bimodal segmentation on a deformed image, which is constructed using the information of the dominant colors in the background.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2011년도 한국컴퓨터종합학술대회논문집 Vol.38 No.1(C)
• /
• pp.365-368
• /
• 2011

흉부 CT 영상은 각종 흉부 질환의 진단에 널리 사용되고 있으며, 영상 분석 기술을 이용하여 암이나 다양한 형태의 종양 등의 관심 영역(region-of-interest)을 자동으로 탐지하는 기법들이 최근 많이 연구되고 있다. 본 논문에서는 흉부 CT 영상에서 연속된 여러 장의 2D 영상의 변화를 분석하여 중요 관심 영역인 노듈(nodule)을 찾는 방법을 제안한다. 인접 영상에서의 노듈의 연속성과 크기의 변화, 모양, 밝기의 특징을 분석하여 노듈을 검출하였고, 실험 결과, 상당한 정도의 정확도를 보였다. 향후 추가적인 연구를 통하여 실제 환경에서도 활용 가능할 것으로 기대된다.