• 한국인터넷방송통신학회논문지
• /
• 제13권2호
• /
• pp.181-189
• /
• 2013

본 논문에서는 패턴 인식기를 설계하기 위하여 개별 입력 공간을 기반으로 한 퍼지 뉴럴 네트워크를 소개한다. 제안된 퍼지 뉴럴 네트워크는 각 입력 공간을 개별적으로 분할함으로서 네트워크를 구성한다. 규칙의 전반부는 개별적 입력 공간을 퍼지 분할하여 독립적으로 구성하고, 규칙의 후반부는 다항식으로서 표현된다. 퍼지 뉴럴 네트워크의 학습은 퍼지 규칙의 후반부에 있는 뉴런의 연결가중치를 조정함으로써 실현되고, 오류 역전파 알고리즘을 이용하여 실현한다. 또한, 제안한 네트워크의 파라미터를 최적화하기 위하여 실수 코딩 유전자 알고리즘을 이용한다. 마지막으로, 패턴 인식을 위한 실험 데이터를 이용하여 최적화된 패턴 인식기를 설계한다.

• 한국산학기술학회논문지
• /
• 제12권11호
• /
• pp.5164-5171
• /
• 2011

비선형 공정에 대한 퍼지 모델링은 일반적으로 주어진 데이터를 이용하여 입력 변수를 선정하고 각 입력 변수에 대한 입력 공간을 분할하여 이들 입력 변수 및 공간 분할에 의해 퍼지 규칙을 형성한다. 퍼지 규칙의 전반부는 입력 변수 선정, 공간 분할 수 및 소속 함수에 의해 동정되고 퍼지 규칙의 후반부는 간략 추론, 선형 추론에 의해 다항식 함수의 형태로 동정된다. 일반적으로 주어진 데이터를 이용한 비선형 공정에 대한 퍼지 규칙의 형성은 차원이 증가할수록 규칙의 수가 지수적으로 증가하는 문제를 가지고 있다. 이를 해결하기 위해 각 입력 공간의 퍼지 분할에 의한 퍼지 규칙을 개별적으로 형성함으로써 복잡한 비선형 공정을 모델링 할 수 있다. 따라서 본 논문에서는 개별적인 입력 공간을 활용하여 퍼지 규칙을 생성한다. 퍼지 규칙의 전반부 파라미터는 입력 데이터의 최소 값과 최대 값을 이용하는 최소-최대 방법을 이용하여 동정되고, 소속 함수는 삼각형, 범종형, 사다리꼴형 소속 함수를 사용한다. 마지막으로, 비선형 공정으로는 널리 이용되는 데이터를 이용하여 시스템 특성 및 성능을 평가한다.

• Journal of Information Processing Systems
• /
• 제5권1호
• /
• pp.5-10
• /
• 2009

Text data has always accounted for a major portion of the world's information. As the volume of information increases exponentially, the portion of text data also increases significantly. Text classification is therefore still an important area of research. LDA is an updated, probabilistic model which has been used in many applications in many other fields. As regards text data, LDA also has many applications, which has been applied various enhancements. However, it seems that no applications take care of the input for LDA. In this paper, we suggest a way to map the input space to a reduced space, which may avoid the unreliability, ambiguity and redundancy of individual terms as descriptors. The purpose of this paper is to show that LDA can be perfectly performed in a "clean and clear" space. Experiments are conducted on 20 News Groups data sets. The results show that the proposed method can boost the classification results when the appropriate choice of rank of the reduced space is determined.

• International journal of advanced smart convergence
• /
• 제1권2호
• /
• pp.30-33
• /
• 2012

This paper propose a new design of fuzzy neural networks based on hard partition to generate the rules of the networks. For this we use hard c-means (HCM) clustering algorithm. The premise part of the rules of the proposed networks is realized with the aid of the hard partition of input space generated by HCM clustering algorithm. The consequence part of the rule is represented by polynomial functions. And the coefficients of the polynomial functions are learned by BP algorithm. The number of the hard partition of input space equals the number of clusters and the individual partitioned spaces indicate the rules of the networks. Due to these characteristics, we may alleviate the problem of the curse of dimensionality. The proposed networks are evaluated with the use of numerical experimentation.

• 전기학회논문지
• /
• 제60권11호
• /
• pp.2157-2164
• /
• 2011

In this paper, we introduce a new category of fuzzy inference system based on Type-2 fuzzy c-means clustering algorithm (T2FCM-based FIS). The premise part of the rules of the proposed model is realized with the aid of the scatter partition of input space generated by Type-2 FCM clustering algorithm. The number of the partition of input space is composed of the number of clusters and the individual partitioned spaces describe the fuzzy rules. Due to these characteristics, we can alleviate the problem of the curse of dimensionality. The consequence part of the rule is represented by polynomial functions with interval sets. To determine the structure and estimate the values of the parameters of Type-2 FCM-based FIS we consider the successive tuning method with generation-based evolution by means of real-coded genetic algorithms. The proposed model is evaluated with the use of numerical experimentation.

• 한국항공우주학회지
• /
• 제47권5호
• /
• pp.364-370
• /
• 2019

본 연구에서는 능동적인 블레이드 제어기법인 개별 블레이드 제어(Individual Blade Control, IBC) 기법을 적용하여 고속비행 시 동축반전 회전익기의 허브 진동하중을 억제하기 위한 최적 제어입력을 탐색하였다. 통합 공탄성 해석 프로그램인 CAMRAD II를 이용하여 동축반전 회전익기인 XH-59A를 모델링하고 다양한 IBC 입력 조건에 대하여 파라미터 연구를 수행하였다. 파라미터 조절 연구를 통하여 허브 진동억제 성능을 구한 결과, 3/rev 가진 주파수의 $0.5^{\circ}$ 진폭에 $300^{\circ}$ 위상각을 갖는 IBC 제어 입력을 적용할 경우 기준 대비 진동 수준이 최대 50% 감소하는 것을 확인하였다. 진동 억제 성능은 후류 간섭에서 보다 자유로운 상부로터에서 6% 가량 하부로터보다 크게 나타났다. 로터의 전진면에서만 IBC 입력를 가진하는 경우에는 조화 가진 입력과 동일한 입력을 가할 경우 진동 수준이 최대 17% 정도 추가적으로 감소하는 것을 확인하였다. 이러한 진동 감소는 전진면만을 대상으로 적은 에너지 투입 비용으로 달성한 특징이 있다.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2003년도 춘계학술대회 논문집 기술교육전문연구회
• /
• pp.90-95
• /
• 2003

For the safety driving of an automobile which is become individual requisites, a new Neural Network algorithm which recognized the load vehicles in real time is proposed. The proposed neural network technique is the real time computation method through the inter-node diffusion. In the network, a node corresponds to a state in the quantized input space. Each node is composed of a processing unit and fixed weights from its neighbor nodes as well as its input terminal. The most reliable algorithm derived for real time recognition of vehicles, is a dynamic programming based algorithm based on sequence matching techniques that would process the data as it arrives and could therefore provide continuously updated neighbor information estimates. Through several simulation experiments, real time reconstruction of the nonlinear image information is processed. 1-D LIPN hardware has been composed and various experiments with static and dynamic signals have been implemented.

• 한국멀티미디어학회논문지
• /
• 제19권9호
• /
• pp.1698-1709
• /
• 2016

These days, anyone can easily product and share their own content through a web service such as blogs and SNS. However, contents are being operated separately because of the space limitation in individual SNS. Therefore, it is hard to search contents efficiently in individual SNS. To solve this problem, this paper propose a "Private Contents Management and Sharing Service with Voluntary Sharing Economy System." The system is in part [Input], [Save] and it provides a way to collect the content that are scattered on the Internet based on the creation of personal index. It also proposes a more systematic content management and sharing by creating and updating the website standard index by introducing an index Coordinator concept. Furthermore in [Use] section, by providing a portion of the index as the primary search results, it avoid unclassified content list which was simply collected by users. In conclusion, unlike previous studies, this system will contribute to the acquisition and management of interspersed content and ultimately contribute to the shared activation by preventing secondary processing and unauthorized processing to the original article.

• 천문학회보
• /
• 제44권1호
• /
• pp.54.2-54.2
• /
• 2019

Removing noise which occurs inevitably when taking image data has been a big concern. There is a way to raise signal-to-noise ratio and it is regarded as the only way, image stacking. Image stacking is averaging or just adding all pixel values of multiple pictures taken of a specific area. Its performance and reliability are unquestioned, but its weaknesses are also evident. Object with fast proper motion can be vanished, and most of all, it takes too long time. So if we can handle single shot image well and achieve similar performance, we can overcome those weaknesses. Recent developments in deep learning have enabled things that were not possible with former algorithm-based programming. One of the things is generating data with more information from data with less information. As a part of that, we reproduced stacked image from single shot image using a kind of deep learning, conditional generative adversarial network (cGAN). r-band camcol2 south data were used from SDSS Stripe 82 data. From all fields, image data which is stacked with only 22 individual images and, as a pair of stacked image, single pass data which were included in all stacked image were used. All used fields are cut in $128{\times}128$ pixel size, so total number of image is 17930. 14234 pairs of all images were used for training cGAN and 3696 pairs were used for verify the result. As a result, RMS error of pixel values between generated data from the best condition and target data were $7.67{\times}10^{-4}$ compared to original input data, $1.24{\times}10^{-3}$. We also applied to a few test galaxy images and generated images were similar to stacked images qualitatively compared to other de-noising methods. In addition, with photometry, The number count of stacked-cGAN matched sources is larger than that of single pass-stacked one, especially for fainter objects. Also, magnitude completeness became better in fainter objects. With this work, it is possible to observe reliably 1 magnitude fainter object.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 B
• /
• pp.546-549
• /
• 1999

For the safty driving of an automobile which is become individual requisites, a new Neural Network algorithm which recognized the load vehicles in real time is proposed. The proposed neural network technique is the real time computation method through the inter-node diffusion. In the network, a node corresponds to a state in the quantized input space. Each node is composed of a processing unit and fixed weights from its neighbor nodes as well as its input terminal. The most reliable algorithm derived for real time recognition of vehicles, is a dynamic programming based algorithm based on sequence matching techniques that would process the data as it arrives and could therefore provide continuously updated neighbor information estimates. Through several simulation experiments, real time reconstruction of the nonlinear image information is processed 1-D LIPN hardware has been composed and various experiments with static and dynamic signals have been implmented.