• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1952-1953
• /
• 2011

본 논문에서는 다항식 기반 Radial Basis Function(RBF)신경회로망(Polynomial based Radial Basis function Neural Network)을 설계하고 이를 n-클래스 패턴 분류 문제에 적용한다. 제안된 다항식기반 RBF 신경회로망은 입력층, 은닉층, 출력층으로 이루어진다. 입력층은 입력 벡터의 값들을 은닉층으로 전달하는 기능을 수행하고 은닉층과 출력층사이의 연결가중치는 상수, 선형식 또는 이차식으로 이루어지며 경사 하강법에 의해 학습된다. Networks의 최종 출력은 연결가중치와 은닉층 출력의 곱에 의해 퍼지추론의 결과로서 얻어진다. 패턴분류기의 최적화는 PSO(Particle Swarm Optimization)알고리즘을 통해 이루어진다. 그리고 제안된 패턴분류기는 실제 얼굴인식 시스템으로 응용하여 직접 CCD 카메라로부터 입력받은 데이터를 영상 보정, 얼굴 검출, 특징 추출 등과 같은 처리 과정을 포함하여 서로 다른 등록인물의 n-클래스 분류 문제에 적용 및 평가되어 분류기로써의 성능을 분석해본다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.5
• /
• pp.401-408
• /
• 2000

In this paper, we proposed a fuzzy controller that founded by the general feedback control with the new adjustment method when it's tuning. The general feedback controller is operated that supply to the plant making the control input multiplying the appropriate gain of controller on the error between the output of the plant and the reference, But proposed feedback fuzzy controller consist of three loops. The inner loop consists of plant and an ordinary feedback controller. The fuzzy inference of controller performed by the outer loops, which is composed of a fuzzy modeling and inference. We can observe that the output of control system converges toward the reference. Also, the behaviour of feedback fuzzy system is converged from the transient. That is, we verified that designed fuzzy controllers was adapted effectively through the experiments in the hydraulic motor system using floating point DSP processor.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.275-275
• /
• 2023

기후변화는 우리의 현실로 다가와 있지만 기후변화로 인해 어떠한 일이 벌어질 것인지는 정확하게 알 수 없는 문제가 있다. 특히 호우의 강도와 지속시간 등은 수문설계에 영향을 미치는 주요한 인자 임에도 불구하고 과학적이고 합리적인 추론이 쉽지 않다. 본 논문에서는 일본에서 대규모 기후 앙상블 모의실험 기반으로 생성된 d4PDF(Data for Policy Decision Making for Future Change)자료 중 시간 단위의 강수앙상블 모의 자료를 이용하여 기상청 서울지점의 강우강도-지속시간-생기빈도 곡선(Intensity-Duration-Frequency Curve; IDF 곡선)의 변화를 추정해 보았다. 이를 위하여 대용량의 자료를 확보하고 서울지점에서의 과거 50년간의 실측자료와 동일기간의 모의자료에 대한 연최대치 계열에 분위사상법을 적용하여 모의자료의 계통적 오차를 소거할 수 있는 함수를 추정하고 이를 이용하여 미래 시나리오에 적용함으로써 지구평균기온 상승에 대응하는 서울관측지점의 IDF 곡선을 추정하여 제시하였다. 추정 결과의 내용은 다양한 요소에 의해 영향을 받는 미래 기후에 대한 내용이라 신뢰성의 평가가 어렵지만 기존의 강우강도에 일률적으로 위험률을 곱하는 방식보다는 좀 더 합리적인 방법이라 생각되며 향후 수문설계 등에 고려될 수도 있을 것이다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.133-135
• /
• 2021

본 논문에서는 목조 문화재 영상에서의 변위 현상 중 하나인 크랙이 발생하는 영역을 감지하기 위한 임베딩 유사도 기반 모델을 제안한다. 우선 변위가 존재하지 않는 정상으로만 구성된 학습 이미지는 사전 학습된 합성 곱 신경망을 통과하여 임베딩 벡터들을 추출한다. 그 이후 임베딩 벡터들을 가지고 정상 클래스에 대한 분포의 파라미터 값을 구한다. 실제 추론 과정에 사용되는 테스트 이미지에 대해서도 마찬가지로 임베딩 벡터를 구한다. 그런 다음 테스트 이미지의 임베딩 벡터와 이전에 구한 정상 클래스를 대표하는 가우시안 분포 정보와의 거리를 계산하여 이상치 맵을 생성하여 최종적으로 변위가 존재하는 영역을 감지한다. 데이터 셋으로는 충주시 근처의 문화재에 방문해서 수집한 목조 문화재 이미지를 가지고 정상 및 비정상으로 구분한 데이터 셋을 사용하였다. 실험 결과 우리가 제안한 임베딩 유사도 기반 모델이 목조 문화재에서 크랙이 발생하는 변위 영역을 잘 감지함을 확인하였다. 이러한 결과로부터 우리가 제안한 방법이 목재 문화재의 크랙 현상에 대한 변위 영역 검출에 있어서 매우 적합함을 보여준다.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.519-524
• /
• 2023

In hashing-based image retrieval, the hash code of a manipulated image is different from the original image, making it difficult to search for the same image. This paper proposes and evaluates a self-supervised deephashing model that generates perceptual hash codes from feature information such as texture, shape, and color of images. The comparison models are autoencoder-based variational inference models, but the encoder is designed with a fully connected layer, convolutional neural network, and transformer modules. The proposed model is a variational inference model that includes a SimAM module of extracting geometric patterns and positional relationships within images. The SimAM module can learn latent vectors highlighting objects or local regions through an energy function using the activation values of neurons and surrounding neurons. The proposed method is a representation learning model that can generate low-dimensional latent vectors from high-dimensional input images, and the latent vectors are binarized into distinguishable hash code. From the experimental results on public datasets such as CIFAR-10, ImageNet, and NUS-WIDE, the proposed model is superior to the comparative model and analyzed to have equivalent performance to the supervised learning-based deephashing model. The proposed model can be used in application systems that require low-dimensional representation of images, such as image search or copyright image determination.

• Knowledge Management Research
• /
• v.21 no.3
• /
• pp.27-44
• /
• 2020

In order to solve the user's information overload problem, recommender systems infer users' preferences and suggest items that match them. The collaborative filtering (CF), the most successful recommendation algorithm, has been improving performance until recently and applied to various business domains. Visual information, such as book covers, could influence consumers' purchase decision making. However, CF-based recommender systems have rarely considered for visual information. In this study, we propose VizNCS, a CF-based deep learning model that uses visual information as additional information. VizNCS consists of two phases. In the first phase, we build convolutional neural networks (CNN) to extract visual features from image data. In the second phase, we supply the visual features to the NCF model that is known to easy to extend to other information among the deep learning-based recommendation systems. As the results of the performance comparison experiments, VizNCS showed higher performance than the vanilla NCF. We also conducted an additional experiment to see if the visual information affects differently depending on the product category. The result enables us to identify which categories were affected and which were not. We expect VizNCS to improve the recommender system performance and expand the recommender system's data source to visual information.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.179-188
• /
• 2023

Recently, fake news disguises the form of news content and appears whenever important events occur, causing social confusion. Accordingly, artificial intelligence technology is used as a research to detect fake news. Fake news detection approaches such as automatically recognizing and blocking fake news through natural language processing or detecting social media influencer accounts that spread false information by combining with network causal inference could be implemented through deep learning. However, fake news detection is classified as a difficult problem to solve among many natural language processing fields. Due to the variety of forms and expressions of fake news, the difficulty of feature extraction is high, and there are various limitations, such as that one feature may have different meanings depending on the category to which the news belongs. In this paper, emotional change patterns are presented as an additional identification criterion for detecting fake news. We propose a model with improved performance by applying a convolutional neural network to a fake news data set to perform analysis based on content characteristics and additionally analyze emotional change patterns. Sentimental polarity is calculated for the sentences constituting the news and the result value dependent on the sentence order can be obtained by applying long-term and short-term memory. This is defined as a pattern of emotional change and combined with the content characteristics of news to be used as an independent variable in the proposed model for fake news detection. We train the proposed model and comparison model by deep learning and conduct an experiment using a fake news data set to confirm that emotion change patterns can improve fake news detection performance.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.9-16
• /
• 2023

Recently, a convolutional neural network (CNN) based system is being developed to overcome the limitations of human resources in the apple quality classification of farmhouse. However, since convolutional neural networks receive only images of the same size, preprocessing such as sampling may be required, and in the case of oversampling, information loss of the original image such as image quality degradation and blurring occurs. In this paper, in order to minimize the above problem, to generate a image patch based graph of an original image and propose a random walk-based positional encoding method to apply the graph transformer model. The above method continuously learns the position embedding information of patches which don't have a positional information based on the random walk algorithm, and finds the optimal graph structure by aggregating useful node information through the self-attention technique of graph transformer model. Therefore, it is robust and shows good performance even in a new graph structure of random node order and an arbitrary graph structure according to the location of an object in an image. As a result, when experimented with 5 apple quality datasets, the learning accuracy was higher than other GNN models by a minimum of 1.3% to a maximum of 4.7%, and the number of parameters was 3.59M, which was about 15% less than the 23.52M of the ResNet18 model. Therefore, it shows fast reasoning speed according to the reduction of the amount of computation and proves the effect.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.19 no.6
• /
• pp.9-15
• /
• 2005

In this research the discharge characteristics of logic gate of the discharge logic gate plasma display panel with the NOT-AND logic function newly designed was analyzed. As for this discharge logic gate a logical output is induced by controlling the voltage between the electrodes using the discharge path. From the experimental result the discharge characteristics of logic gate is influenced by the interrelation of the voltages appling two vertical electrodes. To in the application possibility to large screen PDP, the discharge characteristics by the line resistance of the electrode was evaluated In result it has been inferred that the influence which the drop of voltage by the line resistance of two vertical electrodes exerts on the discharge of the logic gate is minute. Through the experiment, the optimized values of the pulse voltages and the current limitation resistances of each electrode which composed the discharge logic gate were obtained and maximum operation margin of 49[V] was obtained.

• Journal of IKEEE
• /
• v.6 no.1 s.10
• /
• pp.30-39
• /
• 2002

In this paper, we proposed a new circuit construction method that reduces the number of circuit devices of fuzzy controller. Sasaki had defined a new operator to eliminate the divide circuit comparing with the center of gravity method which often using to design the fuzzy controller. In this paper we obtained the more compacted fuzzy controller's circuit by using the proposed definition of fuzzification and defuzzification than using the Sasaki's method and the fuzzification and defuzzification are reverse operation each other. Using these definitions we exhibit the new design method and circuit structure that can eliminate the bounded product(BP) circuit included in Sasaki's circuit. Using the proposed method to level controlling of the water tank, we verified the fuzzy controller's performance by using existent method and proposed method. As a result that are calculated by using the Proposed fuzzy controller to level controlling of the water tank, total numbers of blocks and devices were decreased. If the number of variables and antecedents are Be11ing larger, this method is more efficient.