• 디지털융복합연구
• /
• 제15권3호
• /
• pp.201-209
• /
• 2017

본 논문은 계층적 Convolutional Nerual Network(CNN)을 이용한 스마트폰용 객체 인식 시스템이다. 전체적인 구성은 스마트폰과 서버를 연결하여 서버에서 컨볼루셔널 뉴럴 네트워크로 객체 인식을 하고 수집된 데이터를 매칭시켜 스마트폰으로 객체의 상세정보를 전달하는 방법이다. 또한 계층적 컨볼루셔널 뉴럴 네트워크와 단편적 컨볼루셔널 뉴럴 네트워크와 비교하였다. 계층적 컨볼루셔널 뉴럴 네트워크는 88%, 단편적 컨볼루셔널 뉴럴 네트워크는 73%의 정확도를 가지며 15%p의 성능 향상을 보였다. 이를 기반으로 스마트폰과 방송매체와 연동한 T-Commerce 시장 확장의 가능성을 보여준다. 아울러 방송영상을 시청하면서 Information Retrieval, AR/VR 서비스도 제공 가능하다.

• 한국지능시스템학회논문지
• /
• 제21권1호
• /
• pp.12-18
• /
• 2011

본 논문에서는 RBF 뉴럴 네트워크에서 은닉층 활성함수에 Interval type-2 퍼지개념을 적용한 새로운 RBF 뉴럴 네트워크를 설계하였다. 퍼지 시스템 분야에서 불확실한 정보에 대한 Type-1 퍼지집합의 성능을 보안하고자 Type-2 퍼지집합이 제안되었으며, 멤버쉽함수 안에 다시 멤버쉽함수를 생성함으로써 불확실한 정보를 좀 더 효과적으로 다루고자 하였다. 따라서 본 논문에서는 RBF 뉴럴 네트워크의 은닉층 활성함수에 type-2 퍼지집합의 개념을 적용하여 불확실한 정보에 대한 모델 성능을 개선하고자 하였다. 나아가 연결가중치를 상수항이 아닌 1차식으로 구성된 다항식을 사용하여 최종출력을 입력-출력의 관계식으로 표현하였다. 연결가중치는 기존의 경사하강법(Gradient Descent Method; GDM) 대신 conjugate gradient method(CGM)을 사용하여 파라미터를 동조하고, 은닉층의 활성함수는 공간탐색 진화 알고리즘(Space Search Evolutionary Algorithm; SSEA)을 이용하여 가우시안 함수의 중심점 및 분포상수를 동조하여 모델의 성능을 개선시킨다. 제안된 모델의 성능을 평가하기 위해 가스로 시계열 데이터를 사용하였으며, 결과를 기존 모델과 비교하였다.

• 대한인간공학회:학술대회논문집
• /
• 대한인간공학회 1993년도 추계학술대회논문집
• /
• pp.124-132
• /
• 1993

This study compared two prediction methods-regression and artificial neural network (ANN) on the visual search performance when monitoring a multi-parameter screen with different occurrence frequencies. Under the highlighting condition for the highest occurrence frequency parameter as a search cue, it was found from the requression analysis that variations of mean search time (MST) could be expained almost by three factors such as the number of parameters, the target occurrence frequency of a highlighted parameter, and the highlighted parameter size. In this study, prediction performance of ANN was evaluated as an alternative to regression method. Backpropagation method which was commonly used as a pattern associator was employed to learn a search behavior of subjects. For the case of increased number of parameters and incresed target occurrence frequency of a highlighted parameter, ANN predicted MST's moreaccurately than the regression method (p<0.000). Only the MST's predicted by ANN did not statistically differ from the true MST's. For the case of increased highlighted parameter size. both methods failed to predict MST's accurately, but the differences from the true MST were smaller when predicted by ANN than by regression model (p=0.0005). This study shows that ANN is a good predictor of a visual search performance and can substitute the regression method under certain circumstances.

• 한국지능시스템학회논문지
• /
• 제21권5호
• /
• pp.602-608
• /
• 2011

본 논문에서는, 신경망을 이용한 뉴로 인터페이스 설계를 통해 논홀로노믹 이동 로봇을 제어하는 방법을 제시하였다. 특히, 가상의 마스터-슬레이브 로봇 개념을 이용하여, 부분적으로 안정된 마스터 로봇의 역 동적모델이 피드백-에러 학습법을 적용한 신경망을 통해 온라인으로 획득되도록 하였다. 이 피드백 제어기는 PD 보상기에 기초를 두고 있다. 온라인 학습을 위한 신경망은 입력층이 6개의 입력세포들($x_i$, i=1~6)로 구성되어있으며, 1개의 은닉층에는 2개의 은닉세포($o_j$, j=1~2), 출력층은 2개의 출력세포(${\tau}_k$, k=1~2)로 구성되었고, 신경망의 온라인 학습을 위하여 최소자승법에 의한 오류역전파 알고리즘을 이용하였다. 본 연구에서 개발된 뉴로 인터페이스의 경로추적제어에 관한 성능은 2-wheel 독립구동이 가능한 논홀로노믹 이동 로봇의 시뮬레이션으로 증명하였다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제52권7호
• /
• pp.424-433
• /
• 2003

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제52권7호
• /
• pp.424-424
• /
• 2003

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

• 전기학회논문지
• /
• 제58권2호
• /
• pp.399-406
• /
• 2009

In this study, Polynomial Radial Basis Function Neural Network(pRBFNN) based on Fuzzy Inference System is designed and its parameters such as learning rate, momentum coefficient, and distributed weight (width of RBF) are optimized by means of Particle Swarm Optimization. The proposed model can be expressed as three functional module that consists of condition part, conclusion part, and inference part in the viewpoint of fuzzy rule formed in 'If-then'. In the condition part of pRBFNN as a fuzzy rule, input space is partitioned by defining kernel functions (RBFs). Here, the structure of kernel functions, namely, RBF is generated from HCM clustering algorithm. We use Gaussian type and Inverse multiquadratic type as a RBF. Besides these types of RBF, Conic RBF is also proposed and used as a kernel function. Also, in order to reflect the characteristic of dataset when partitioning input space, we consider the width of RBF defined by standard deviation of dataset. In the conclusion part, the connection weights of pRBFNN are represented as a polynomial which is the extended structure of the general RBF neural network with constant as a connection weights. Finally, the output of model is decided by the fuzzy inference of the inference part of pRBFNN. In order to evaluate the proposed model, nonlinear function with 2 inputs, waster water dataset and gas furnace time series dataset are used and the results of pRBFNN are compared with some previous models. Approximation as well as generalization abilities are discussed with these results.

• 전기학회논문지
• /
• 제62권4호
• /
• pp.544-553
• /
• 2013

In this paper, we propose the pattern classifier of Radial Basis Function Neural Networks(RBFNNs) for diagnosis of 3-phase partial discharge. Conventional methods map the partial discharge/noise data on 3-PARD map, and decide whether the partial discharge occurs or not from 3-phase or neutral point. However, it is decided based on his own subjective knowledge of skilled experter. In order to solve these problems, the mapping of data as well as the classification of phases are considered by using the general 3-PARD map and PA method, and the identification of phases occurring partial discharge/noise discharge is done. In the sequel, the type of partial discharge occurring on arbitrary random phase is classified and identified by fuzzy clustering-based polynomial Radial Basis Function Neural Networks(RBFNN) classifier. And by identifying the learning rate, momentum coefficient, and fuzzification coefficient of FCM fuzzy clustering with the aid of PSO algorithm, the RBFNN classifier is optimized. The virtual simulated data and the experimental data acquired from practical field are used for performance estimation of 3-phase partial discharge pattern classifier.

• 전기학회논문지
• /
• 제60권12호
• /
• pp.2352-2360
• /
• 2011

In this paper, we introduce a comprehensive design methodology of Radial Basis Function Neural Networks (RBFNN) that is based on mechanism of clustering and optimization algorithm. We can divide some clusters based on similarity of input dataset by using clustering algorithm. As a result, the number of clusters is equal to the number of nodes in the hidden layer. Moreover, the centers of each cluster are used into the centers of each receptive field in the hidden layer. In this study, we have applied Fuzzy-C Means(FCM) and K-Means(KM) clustering algorithm, respectively and compared between them. The weight connections of model are expanded into the type of polynomial functions such as linear and quadratic. In this reason, the output of model consists of relation between input and output. In order to get the optimal structure and better performance, Particle Swarm Optimization(PSO) is used. We can obtain optimized parameters such as both the number of clusters and the polynomial order of weights connection through structural optimization as well as the widths of receptive fields through parametric optimization. To evaluate the performance of proposed model, NXT equipment offered by National Instrument(NI) is exploited. The situation awareness system-related intelligent model was built up by the experimental dataset of distance information measured between object and diverse sensor such as sound sensor, light sensor, and ultrasonic sensor of NXT equipment.