• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1991년도 춘계학술대회 논문집
• /
• pp.225-236
• /
• 1991

Thermoleastoplastic analysis of a typicaldevice consisting of tube and moving part having an initial clearance in between, which is subjected mechanical and thermalload, has been carried out toexamine the cause of malfunction mainly at hightemperature condition, and to improve the design. Permant deformation, extraction force, and stress distributions are discussed by using finite element method. The extraction energy can be remarkably reduced by changing the configuration of the cylinder from straight-type to ta[er-type and this effect has been proved by sustained cyclic test.

• 제어로봇시스템학회논문지
• /
• 제16권12호
• /
• pp.1150-1158
• /
• 2010

In this study, the Polynomial-based Radial Basis Function Neural Networks is proposed as one of the recognition part of overall face recognition system that consists of two parts such as the preprocessing part and recognition part. The design methodology and procedure of the proposed pRBFNNs are presented to obtain the solution to high-dimensional pattern recognition problem. First, in preprocessing part, we use a CCD camera to obtain a picture frame in real-time. By using histogram equalization method, we can partially enhance the distorted image influenced by natural as well as artificial illumination. We use an AdaBoost algorithm proposed by Viola and Jones, which is exploited for the detection of facial image area between face and non-facial image area. As the feature extraction algorithm, PCA method is used. In this study, the PCA method, which is a feature extraction algorithm, is used to carry out the dimension reduction of facial image area formed by high-dimensional information. Secondly, we use pRBFNNs to identify the ID by recognizing unique pattern of each person. The proposed pRBFNNs architecture consists of three functional modules such as the condition part, the conclusion part, and the inference part as fuzzy rules formed in 'If-then' format. In the condition part of fuzzy rules, input space is partitioned with Fuzzy C-Means clustering. In the conclusion part of rules, the connection weight of pRBFNNs is represented as three kinds of polynomials such as constant, linear, and quadratic. Coefficients of connection weight identified with back-propagation using gradient descent method. The output of pRBFNNs model is obtained by fuzzy inference method in the inference part of fuzzy rules. The essential design parameters (including learning rate, momentum coefficient and fuzzification coefficient) of the networks are optimized by means of the Particle Swarm Optimization. The proposed pRBFNNs are applied to real-time face recognition system and then demonstrated from the viewpoint of output performance and recognition rate.

• 대한전자공학회논문지SP
• /
• 제41권4호
• /
• pp.45-50
• /
• 2004

본 논문에서는 실시간 얼굴인식 시스템을 위한 새로운 PCA 기반 LDA 혼합 알고리즘을 제안한다. 크게 얼굴추출 부분과 얼굴인식 부분으로 구성되어 있으며, 얼굴추출 부분에는 차영상, color filtering, 눈과 입의 영역 검출 그리고 정규화 방법을 사용하였고, 얼굴인식 부분에는 추출된 얼굴 후보 영역 영상에 PCA와 LDA를 혼합하여 적용하였다. 기존의 PCA만을 사용한 인식시스템은 낮은 인식률을 보였으며, LDA만을 사용한 인식시스템에서는 학습데이터의 수에 비하여 영상의 화소 개수가 많은 경우 LDA를 입력 영상에 그대로 적용하기 곤란하였다. 이러한 단점을 극복하기 위하여, 정규화 된 영상에 PCA를 적용하여 차원을 축소한 후 LDA를 사용하여 실시간 인식을 가능하게 하였으며, 인식률 또한 향상시킬 수 있었다. 제안한 시스템의 성능을 평가하기 위하여 자체 제작한 DAUface의 데이터베이스를 가지고 실험을 하였다. 실험 결과, 제안된 방법이 PCA 방법과 LDA 방법, 그리고 ICA 방법에 비해 인식률이 상당히 우수함을 알 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1639-1658
• /
• 2019

Named entity recognition (NER) is an important technique for improving the performance of data mining and big data analytics. In previous studies, NER systems have been employed to identify named-entities using statistical methods based on prior information or linguistic features; however, such methods are limited in that they are unable to recognize unregistered or unlearned objects. In this paper, a method is proposed to extract objects, such as technologies, theories, or person names, by analyzing the collocation relationship between certain words that simultaneously appear around specific words in the abstracts of academic journals. The method is executed as follows. First, the data is preprocessed using data cleaning and sentence detection to separate the text into single sentences. Then, part-of-speech (POS) tagging is applied to the individual sentences. After this, the appearance and collocation information of the other POS tags is analyzed, excluding the entity candidates, such as nouns. Finally, an entity recognition model is created based on analyzing and classifying the information in the sentences.

• Journal of Information Processing Systems
• /
• 제15권4호
• /
• pp.931-956
• /
• 2019

Documents contain information that can be used for various applications, such as question answering (QA) system, information retrieval (IR) system, and recommendation system. To use the information, it is necessary to develop a method of extracting such information from the documents written in a form of natural language. There are several kinds of the information (e.g., temporal information, spatial information, semantic role information), where different kinds of information will be extracted with different methods. In this paper, the existing studies about the methods of extracting the temporal information are reported and several related issues are discussed. The issues are about the task boundary of the temporal information extraction, the history of the annotation languages and shared tasks, the research issues, the applications using the temporal information, and evaluation metrics. Although the history of the tasks of temporal information extraction is not long, there have been many studies that tried various methods. This paper gives which approach is known to be the better way of extracting a particular part of the temporal information, and also provides a future research direction.

• 대한금속재료학회지
• /
• 제47권5호
• /
• pp.297-303
• /
• 2009

Solvent extraction experiments of cupric and cuprous chloride with Alamine336 have been performed from HCl solution. In order to identify the solvent extraction reaction, distribution diagram of cupric and cuprous species with HCl concentration was obtained by considering complex formation reaction and the activity coefficient of solutes with Pitzer equation. Analysis of the solvent extraction data by graphical method together with the distribution diagram of copper indicated that solvent extraction reaction of copper with Alamine336 depends on HCl concentration. In strong HCl solution of 3 and 5 M, ${CuCl_4}^{2-}$ and ${CuCl_3}^{2-}$ took part in the solvent extraction reaction as Cu(II) and Cu(I), respectively. When HCl concentration was 1 M, ${CuCl_2}^-$ was extracted into the organic phase in the case of Cu(I) while adduct formation between $Cu^{2+}$ and Alamine336 was responsible for the solvent extraction reaction of Cu(II).

• 대한음성학회지:말소리
• /
• 제66호
• /
• pp.87-103
• /
• 2008

We describe a new active learning method in conditional random fields (CRFs) framework for keyphrase extraction. To save elaboration in annotation, we use diversity and representative measure. We select high diversity training candidates by sentence confidence value. We also select high representative candidates by clustering the part-of-speech patterns of contexts. In the experiments using dialog corpus, our method achieves 86.80% and saves 88% training corpus compared with those of supervised method. From the results of experiment, we can see that the proposed method shows improved performance over the previous methods. Additionally, the proposed method can be applied to other applications easily since its implementation is independent on applications.

• 한국약용작물학회지
• /
• 제21권6호
• /
• pp.486-492
• /
• 2013

This study has been conducted to establish the optimal extraction process and HPLC analysis method for the determination of marker compounds as a part of the materials standardization for the development of health functional food materials from Astragali radix. Five extraction conditions including the shaking extraction at room temperature and the reflux extraction at $85^{\circ}C$ with 30%, 50% and 95% ethanol were evaluated. Reflux extraction with 50% ethanol showed the highest extraction yield as $27.27{\pm}2.27%$, while the extraction under reflux with 95% ethanol showed significantly the lowest yield of $10.55{\pm}0.24%$. The quantitative determination methods of calycosin-7-O-${\beta}$-D-glucoside and calycosin as marker compounds of Astragali radix extracts were optimized by HPLC analysis using a Thermo Hypersil column ($4.6{\times}250mm$, $5{\mu}m$) with the gradient elution of water and acetonitrile as the mobile phase at the flow rate of $0.8mLmin^{-1}$ and a detection wavelength of 230nm. The HPLC/UV method was applied successfully to the quantification of two marker compounds in Astragali radix extracts after validation of the method with the linearity, accuracy and precision. The contents of calycosin-7-O-${\beta}$-D-glucoside and calycosin in 50% ethanol extracts by reflux extraction were significantly higher as $1,700.3{\pm}30.4$ and $443.6{\pm}8.4{\mu}g-1$, respectively, comparing with those in other extracts. The results indicate that the reflux extraction with 50% ethanol at $85^{\circ}C$ is optimal for the extraction of Astragali radix, and the established HPLC method are very useful for the evaluation of marker compounds in Astragali radix extracts to develop the health functional material from Astragali radix.

• 한국멀티미디어학회논문지
• /
• 제13권4호
• /
• pp.504-511
• /
• 2010

본 논문에서는 인간 뇌의 내부에 존재하는 해마를 모델링한 해마 신경망을 사용하여 도시방범용 CCTV를 위한 얼굴영역 인식 시스템을 제안한다. 이 시스템은 특징추출 부분과 학습 및 인식 부분으로 구성되어 있으며, 특징 추출 부분은 PCA(Principal Component Analysis)와 LDA(Linear Discriminant Analysis) 사용하여 구성한다. 학습부분에서는 해마의 구조의 순서에 따라 입력되는 영상 데이터들의 특징을 치아 이랑 영역에서 호감도 조정에 의해 반응 패턴을 이진화 하고, 다음으로 CA3 영역에서의 자기 연상을 통해 영상에 포함되어 있는 노이즈를 제거하게 된다. 노이즈가 제거된 데이터는 CA1 영역에서 신경망을 통해 장기기억이 이루어진다. 제안한 시스템의 성능을 평가하기 위해 형태변화와 조명변화에 따른 인식률 실험을 실시하였다. 실험 결과, 본 논문에서 제안한 특징 추출 및 학습 방법을 다른 학습 방법들과 비교하였을 때, 우수한 인식률을 가짐을 확인하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1999년도 제14차 학술회의논문집
• /
• pp.262-265
• /
• 1999

This paper describes a new method for object extraction necessary for image tracking systems. The extraction method which this paper proposes here is that an M-array is set between a camera and the object and the obtained image including the object and M-array is pro-cessed for extracting the object. The image processing utilizes a characteristic of M-array which is robust to noise. When an M-array is overlapped on the object in background image, the object woud have a part of M-array, which is detected by use of partial correlation between the mosaic image of M-array and the standard M-array. Thus the shape and position of the object are extracted by extracting a common domain of width of high correlation value. Experiments are carried out by using an actual photo of Kumamoto city taken from an airplane as background, and by use of a rectangular and circular object. The results of experiment show a wide application of this method for practical image tracking systems.