• Journal of Information Processing Systems
• /
• 제16권4호
• /
• pp.870-881
• /
• 2020

Because SCADA monitoring data of wind turbines are large and fast changing, the unbalanced proportion of data in various working conditions makes it difficult to process fault feature data. The existing methods mainly introduce new and non-repeating instances by interpolating adjacent minority samples. In order to overcome the shortcomings of these methods which does not consider boundary conditions in balancing data, an improved over-sampling balancing algorithm SC-SMOTE (safe circle synthetic minority oversampling technology) is proposed to optimize data sets. Then, for the balanced data sets, a fault diagnosis method based on improved k-nearest neighbors (kNN) classification for wind turbine blade icing is adopted. Compared with the SMOTE algorithm, the experimental results show that the method is effective in the diagnosis of fan blade icing fault and improves the accuracy of diagnosis.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.1738-1764
• /
• 2019

With the advent of big data, deep learning technology has become an important research direction in the field of machine learning, which has been widely applied in the image processing, natural language processing, speech recognition and online advertising and so on. This paper introduces deep learning techniques from various aspects, including common models of deep learning and their optimization methods, commonly used open source frameworks, existing problems and future research directions. Firstly, we introduce the applications of deep learning; Secondly, we introduce several common models of deep learning and optimization methods; Thirdly, we describe several common frameworks and platforms of deep learning; Finally, we introduce the latest acceleration technology of deep learning and highlight the future work of deep learning.

• 한국정밀공학회지
• /
• 제13권9호
• /
• pp.62-69
• /
• 1996

Interpolator is a very important element in NC machines in that it controls tool path and speed. In this paper, studied were extensive interpolation characteristics of reference pulse method among various interpolation and pulse generation methods. Specifically, processing speed and path error of DDA, SPD and SFG methods were compared and analyzed against line, circle and elipse. As a result, in the point of processing speed, SPD method was found to be the best for line interpolation, SFG method for circle and ellipse, and DDA method was found to be the slowest for all paths. In the point of path error, DDA method was found to have the biggest error for all kinds of paths.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.493-495
• /
• 2023

Deep learning methods and attention mechanisms have been incorporated to improve facial emotion recognition, which has recently attracted much attention. The fusion approaches have improved accuracy by combining various types of information. This research proposes a fusion network with self-attention and local attention mechanisms. It uses a multi-layer perceptron network. The network extracts distinguishing characteristics from facial images using pre-trained models on RAF-DB dataset. We outperform the other fusion methods on RAD-DB dataset with impressive results.

• 한국컴퓨터정보학회논문지
• /
• 제26권7호
• /
• pp.19-27
• /
• 2021

본 논문에서는 영상 속 인물을 탐지하고 얼굴 영역을 검출하는 방법을 제안하며, 이 방법은 2가지 작업으로 구성한다. 첫째, 서로 다른 두 명의 인물을 구분하여 프레임 내 인물의 얼굴 위치를 탐지한다. 빠른 탐지를 위해 영상 내 물체를 실시간으로 검출하는 YOLO(You Only Look Once)를 이용하여 얼굴의 위치를 탐지하고 객체탐지상자로 나타낸다. 둘째, 객체탐지상자를 바탕으로 정확한 얼굴 면적을 검출하기 위해 3가지 영상처리 방법을 제시한다. 각 방법은 검출 도형으로 추정한 영역에서 추출한 HSV 값을 이용하여 인물의 얼굴 영역을 검출하였으며 검출 도형의 크기와 모양을 바꾸어 각 방법의 정확도를 비교하였다. 각 얼굴 검출 방법은 신뢰성 검증을 위해 비교 데이터와 영상처리 데이터로 비교 및 분석하였다. 그 결과 원형, 직사각형, 분할 직사각형 방법 중 분할된 직사각형 방법을 사용했을 때 87%로 가장 높은 정확도를 달성하였다.

• 융합신호처리학회논문지
• /
• 제11권2호
• /
• pp.92-96
• /
• 2010

디인터레이싱 방식은 일반적으로 화면 간 디인터레이싱과 화면 내 디인터레이싱으로 나뉜다. 화면 내 디인터레이싱에서 가장 보편적인 방식은 라인카피와 라인별 평균값을 취하는 선형적인 방식이다. 또한 ELA(Edge Based Line Average), ELA를 변형시킨 방식 등이 에지를 이용한 디인터레이싱 방식들이다. 하지만 선형적 보간법은 에지 블러링이나 계단현상을 발생시키고, ELA나 변형된 ELA를 이용한 방식은 특정방향에지 정보가 부족해서 다양한 형태의 에지 부분에서의 디인터레이싱이 좋지 못하다. 본 논문은 7방향의 에지 방향 보간법과 수평방향의 에지라인을 고려한 화면 내 디인터레이싱 방식을 소개한다. 기존 방식에서 나타나던 오브젝트의 계단현상을 최대한 줄여줌으로써 보다 낳은 화질을 보인다.

• 응용통계연구
• /
• 제24권6호
• /
• pp.1103-1113
• /
• 2011

Gene expression data is obtained through many stages of an experiment and errors produced during the process may cause missing values. Due to the distinctness of the data so called 'small n large p', genes have to be selected for statistical analysis, like classification analysis. For this reason, imputation and gene selection are important in a microarray data analysis. In the literature, imputation, gene selection and classification analysis have been studied respectively. However, imputation, gene selection and classification analysis are sequential processing. For this aspect, we compare the performance of classification methods after imputation and gene selection methods are applied to microarray data. Numerical simulations are carried out to evaluate the classification methods that use various combinations of the imputation and gene selection methods.

• 한국환경농학회지
• /
• 제36권1호
• /
• pp.7-16
• /
• 2017

BACKGROUND: Increased value added by rapeseed (Brassica napus) by-product and the development of a usable rapeseed functional tea. METHODS AND RESULTS: To develop a usable rapeseed functional tea, the total flavonoid content in the varieties Youngsan, Tammi, Tamra, Naehan, Hanra, Mokpo No. 68, and Mokpo No. 111 was investigated. Effect of three treatments, i.e., drying, leaching, and roasting, on flavonoid contents or flower was tested using multiple processing methods per treatment. Total flavonoid content decreased under the various drying methods, confirming that flavonoid content is heat-dependent. This finding was more pronounced for freezing and oven-drying (15.3 and 13.8 mg/g DW, respectively), with a 10% difference in the total flavonoid content between the two methods. Under leaching conditions, the flavonoid content decreased with increasing treatment time. Notably, roasting methods did not result in loss of flavonoid content. The total flavonoid content in the rapeseed varieties decreased in the following order: Youngsan, Tammi, Tamra, Naehan, Hanra, Mokpo No. 68, and Mokpo No. 111. CONCLUSION: The flavonoid content in rapeseed flower was higher in Youngsan than in the other varieties, under processing conditions such as freeze-drying, leaching at $90^{\circ}C$ for 5 min, and roasting.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2010년도 춘계학술대회
• /
• pp.441-443
• /
• 2010

다양한 센서들로부터 실시간으로 수집되는 데이터를 분석, 처리해야 하는 유비쿼터스 환경의 다양한 응용 서비스를 개발하기 위해서는 반드시 연속질의 처리 방법이 요구된다. 기존 연구에서 연속질의는 대부분 간격질의와 영역질의만으로 표현되고, 이러한 질의에 대한 처리 기법들이 제시되었다. 그러나 연속질의의 종류는 매우 다양하며, 속성 조건, 공간 조건, 시간 조건의 결합을 통해 표현될 수 있다. 이 논문에서는 연속질의의 종류를 분류하고, 여러 조건들의 결합을 통해 표현 가능한 연속질의 모델을 제안하고자 한다. 이 논문은 연속질의를 표현하는 질의 모델을 제안함으로써, 연속질의 처리를 위한 다양한 후속 연구에 대한 방향을 제시하는데 기여하였다.

• 한국정보처리학회논문지
• /
• 제6권11호
• /
• pp.3140-3149
• /
• 1999

Skew correction for document images can be using a rotational transformation of pixel coordinates. In this paper we propose a method which corrects the document skew, by an amount of $\theta$ degrees, using block information, where the block is defined as a rectangular area containing adjacent black pixels. Processing speed of the proposed method is faster than that of the method using pixel transformation, since the number of floating-point operations can be reduced significantly. In the proposed method, we rotate only the four corner points of each block, and then identify the pixels inside the block. Two methods for inside pixel identification are proposed; the first method finds two points intersecting the boundary of the rotated block in each row, and determines the pixels between the two intersection points as the inside pixel. The second method finds boundary points based on Bresenham's line drawing algorithm, using fixed-point operation, and fills the region surrounded by these boundaries as black pixels. We have measured the performance of the proposed method by experimenting it with 2,016 images of various English and Korean documents. We have also proven the superiority of our algorithm through performance comparison with respect to existing methods based on pixel transformation.