• 한국신재생에너지학회:학술대회논문집
• /
• 한국신재생에너지학회 2009년도 춘계학술대회 논문집
• /
• pp.554-556
• /
• 2009

Wind classification for exact estimation of wind energy resources was carried out using numerically simulated wind data for three years. The MM5(a fifth-generation Mesoscale Model), developed at Penn State University and the National Center for Atmospheric Research (NCAR), was used to estimate the wind fields in this study. We also use a variant of the K-mean clustering to classify the wind district and define the relation between districts. Wind estimated at surface and 100 m high at Busan area is classified into the 10 and 7 classes, respectively. These discrepancies of wind districts pattern at surface and upper air meteorological data indicates the quantity of wind resources can be changed according to the level of wind data used in estimation. Therefore, the estimation of wind district classification by reasonable wind data is utilized to build the effective policy for wind energy dissemination.

• 응용통계연구
• /
• 제29권1호
• /
• pp.193-203
• /
• 2016

본 논문은 ICT기반 시장에서의 수요관리시스템에서의 핵심 요소인 전력 수요 예측을 위하여, 전체 사용량을 기반으로 예측 하는 방식이 아닌, 시계열 기반 군집분석을 통한 군집별 예측량의 결합을 실시하였다. 시계열 군집 분석 방법으로서 Periodogram 기반의 정규화 군집분석, 예측 기반의 군집분석, DTW(Dynamic Time Warping)를 이용하여 군집화를 시도하였으며, 군집 별 수요예측 모형으로서 DSHW(Double Seasonal Holt-Winters) 모형, TBATS(Trigonometric, Box-Cox transform, ARMA errors, Trend and Seasonal components) 모형, FARIMA(Fractional ARIMA) 모형을 사용하여 예측을 실시하였다. 전체 사용량을 기반으로 예측 하는 방식이 아닌, 군집분석을 통한 군집별 예측량의 결합이 더 낮은 MAPE로 나타남에 따라 우수한 예측 방법으로 판단되었다.

• 방송공학회논문지
• /
• 제21권3호
• /
• pp.425-435
• /
• 2016

본 논문에서는 카메라로부터 입력된 영상으로부터 쌀, 커피, 녹차 등 다양한 원료를 양품과 불량품으로 자동 분류하기 위한 분류 모델을 제안한다. 현재 농산물 원료 분류를 위해서 주로 숙달된 노동력의 육안 선택에 의존하고 있지만 작업시간이 길어질수록 반복적인 작업에 의해 분류 능력이 현저히 떨어지는 문제점이 있다. 노동력에 부분적으로 의존하는 기존 제품의 문제점을 해결하기 위해, 본 논문에서는 평균-이동 클러스터링 알고리즘과 단계별 영역 병합 알고리즘을 결합하는 비전기반 자동 원료 분류 알고리즘을 제안한다. 우선 입력 원료 영상에서 평균-이동 클러스터링 알고리즘을 적용하여 영상을 N개의 클러스터 영역으로 분할한다. 다음단계에서 N개의 클러스터 영역 중에서 대표 영역을 선택하고 이웃 영역들의 영역의 색상과 위치 근접성을 기반으로 단계별 영역 병합 알고리즘을 적용하여 유사한 클러스터 영역을 병합한다. 병합된 원료 객체는 RG, GB, BR의 2D 색상 분표로 표현되고, 병합된 원료 객체에 대해 색상 분포 타원을 만든다. 이후 미리 실험적으로 설정된 임계값을 적용하여 원료를 양품과 불량품을 구분한다. 다양한 원료 영상에 대해 본 논문에서 제안하는 알고리즘을 적용한 결과 기존의 클러스터링 알고리즘이나 상업용 분류 방법에 비해 사용자의 인위적 조작이 덜 필요하고 분류성능이 우수한 결과를 나타냄을 알 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제3권2호
• /
• pp.134-146
• /
• 2009

Traffic classification seeks to assign packet flows to an appropriate quality of service(QoS) class based on flow statistics without the need to examine packet payloads. Classification proceeds in two steps. Classification rules are first built by analyzing traffic traces, and then the classification rules are evaluated using test data. In this paper, we use self-organizing map and K-means clustering as unsupervised machine learning methods to identify the inherent classes in traffic traces. Three clusters were discovered, corresponding to transactional, bulk data transfer, and interactive applications. The K-nearest neighbor classifier was found to be highly accurate for the traffic data and significantly better compared to a minimum mean distance classifier.

• Journal of Mechanical Science and Technology
• /
• 제18권12호
• /
• pp.2137-2147
• /
• 2004

The recycling cell formation problem means that disposal products are classified into recycling part families using group technology in their end-of-life phase. Disposal products have the uncertainties of product status by usage influences during product use phase, and recycling cells are formed design, process and usage attributes. In order to deal with the uncertainties, fuzzy set theory and fuzzy logic-based neural network model are applied to recycling cell formation problem for disposal products. Fuzzy C-mean algorithm and a heuristic approach based on fuzzy ART neural network is suggested. Especially, the modified Fuzzy ART neural network is shown that it has a good clustering results and gives an extension for systematically generating alternative solutions in the recycling cell formation problem. Disposal refrigerators are shown as examples.

• 한국멀티미디어학회논문지
• /
• 제18권2호
• /
• pp.138-148
• /
• 2015

Object detection methods based on background learning are widely used in video surveillance. However, when a car runs with headlights on, these methods are likely to detect the car region and the area illuminated by the headlights as one connected change region. This paper describes a method of separating the car region from the area illuminated by the headlights. First, we detect change regions with a background learning method, and extract blobs, connected components in the detected change region. If a blob is larger than the maximum object size, we extract candidate object regions from the blob by clustering the intensity histogram of the frame difference between the mean of background images and an input image. Finally, we compute the similarity between the mean of background images and the input image within each candidate region and select a candidate region with weak similarity as an object region.

• 디지털융복합연구
• /
• 제11권12호
• /
• pp.459-465
• /
• 2013

본 논문은 영상 속 객체와 배경의 컬러 특징을 이용한 주요 객체의 자동 추출 방법에 관한 연구이다. 인간이 객체를 판단할 때에는 배경과 객체의 색상 차이를 이용하는데 이러한 요소를 객체 추출 방법에 적용시키기 위해서는 배경과 객체의 색차를 강조하여야 한다. 따라서 본 논문에서는 원 RGB 영상을 인간의 시각 시스템과 유사한 HSV 색 공간으로 변환하고 각기 다른 분포도의 메디안 필터를 적용한 두 개의 영상을 생성한 뒤 두 개의 메디안 필터가 적용된 영상들을 합산하였고 데이터 군집화 방법인 Mean Shift 알고리즘을 적용하여 색상 특징을 그룹화 하였다. 마지막으로 이진화 작업을 위하여 영상의 채널 수를 3 채널에서 1 채널로 정규화 한 뒤 영상 내 픽셀들의 평균값을 임계값으로 이용하는 이진화 방법으로 객체 지도 영상을 생성하였고 주요 객체를 추출하였다.

• 한국태양에너지학회 논문집
• /
• 제36권1호
• /
• pp.11-18
• /
• 2016

The local wind systems in the Pohang region were categorized into wind sectors. Still, thorough knowledge of wind resource assessment, wind environment analysis, and atmospheric environmental impact assessment was required since the region has outstanding wind resources, it is located on the path of typhoon, and it has large-scale atmospheric pollution sources. To overcome the resolution limitation of meteorological dataset and problems of categorization criteria of the preceding studies, the high-resolution wind resource map of the Korea Institute of Energy Research was used as time-series meteorological data; the 2-step method of determining the clustering coefficient through hierarchical clustering analysis and subsequently categorizing the wind sectors through non-hierarchical K-means clustering analysis was adopted. The similarity of normalized time-series wind vector was proposed as the Euclidean distance. The meteor-statistical characteristics of the mean vector wind distribution and meteorological variables of each wind sector were compared. The comparison confirmed significant differences among wind sectors according to the terrain elevation, mean wind speed, Weibull shape parameter, etc.

• 한국멀티미디어학회논문지
• /
• 제20권12호
• /
• pp.1951-1959
• /
• 2017

Due to the fact that 3D printing is applied to many areas of life, 3D printing models are often used illegally without any permission from the original providers. This paper presents a novel watermarking algorithm for the copyright protection and ownership identification for 3D printing based on the radius curvature of 3D triangle. 3D triangles are extracted and classified into groups based on radius curvature by the clustering algorithm, and then the mean radius curvature of each group will be computed for watermark embedding. The watermark data is embedded to the groups of 3D triangle by changing the mean radius curvature of each group. In each group, we select a 3D triangle which has the nearest radius curvature with the changed mean radius curvature. Finally, we change the vertices of the selected facet according to the changed radius curvature has been embedded watermark. In experiments, the distance error between the original 3D printing model and the watermarked 3D printing model is approximate zero, and the Bit Error Rate is also very low. From experimental results, we verify that the proposed algorithm is invisible and robustness with geometric attacks rotation, scaling and translation.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2007년도 한국음성과학회 공동학술대회 발표논문집
• /
• pp.33-36
• /
• 2007

In this paper, we investigate a fast speaker adaptation method based on eigenvoice in several noisy environments. In order to overcome its weakness against noise, we propose a noisy environment clustering method which divides the noisy adaptation utterances into utterance groups with similar environments by the vector quantization based clustering using a cepstral mean as a feature vector. Then each utterance group is used for adaptation to make an environment dependent model. According to our experiment, we obtained 19-37 % relative improvement in error rate compared with the simultaneous speaker adaptation and environmental compensation method