• The KIPS Transactions:PartB
• /
• v.10B no.7
• /
• pp.777-784
• /
• 2003

In this Paper. we propose a recognition method of machine printed characters based on the combination of recognition units using multiple neural networks. In our recognition method, the input character is classified into one of 7 character types among which the first 6 types are for Hangul character and the last type is for non-Hangul characters. Hangul characters are recognized by several MLP (multilayer perceptron) neural networks through two stages. In the first stage, we divide Hangul character image into two or three recognition units (HRU : Hangul recognition unit) according to the combination fashion of graphemes. Each recognition unit composed of one or two graphemes is recognized by an MLP neural network with an input feature vector of pixel direction angles. In the second stage, the recognition aspect features of the HRU MLP recognizers in the first stage are extracted and forwarded to a subsequent MLP by which final recognition result is obtained. For the recognition of non-Hangul characters, a single MLP is employed. The recognition experiments had been performed on the character image database collected from 50,000 real letter envelope images. The experimental results have demonstrated the superiority of the proposed method.

• Advances in materials Research
• /
• v.6 no.2
• /
• pp.169-184
• /
• 2017

Electro-Discharge machining (EDM) is a thermal process comprising a complex metal removal mechanism. This method works by forming of a plasma channel between the tool and the workpiece electrodes leading to the melting and evaporation of the material to be removed. EDM is considered especially suitable for machining complex contours with high accuracy, as well as for materials that are not amenable to conventional removal methods. However, several phenomena can arise and adversely affect the surface integrity of EDMed workpieces. These have to be taken into account and studied in order to optimize the process. Recently, artificial neural networks (ANN) have emerged as a novel modeling technique that can provide reliable results and readily, be integrated into several technological areas. In this paper, we use an ANN, namely, the multi-layer perceptron and the back propagation network (BPNN) to predict the mean surface roughness of electro-discharge machined surfaces. The comparison of the derived results with experimental findings demonstrates the promising potential of using back propagation neural networks (BPNNs) for getting a reliable and robust approximation of the Surface Roughness of Electro-discharge Machined Components.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2519-2521
• /
• 2003

In this paper, we propose Self-Organizing Fuzzy Polynomial Neural Networks(SOFPNN) architecture for optimal model identification and discuss a comprehensive design methodology supporting its development. It is shown that this network exhibits a dynamic structure as the number of its layers as well as the number of nodes in each layer of the SOFPNN are not predetermined (as this is the case in a popular topology of a multilayer perceptron). As the form of the conclusion part of the rules, especially the regression polynomial uses several types of high-order polynomials such as linear, quadratic, and modified quadratic. As the premise part of the rules, both triangular and Gaussian-like membership function are studied and the number of the premise input variables used in the rules depends on that of the inputs of its node in each layer. We introduce two kinds of SOFPNN architectures, that is, the basic and modified one with both the generic and the advanced type. The superiority and effectiveness of the proposed SOFPNN architecture is demonstrated through nonlinear function numerical example.

• Computers and Concrete
• /
• v.22 no.4
• /
• pp.355-363
• /
• 2018

In the following paper, a socio-political heuristic search approach, named the imperialist competitive algorithm (ICA) has been used to improve the efficiency of the multi-layer perceptron artificial neural network (ANN) for predicting the compressive strength of concrete. 173 concrete samples have been investigated. For this purpose the values of slump flow, the weight of aggregate and cement, the maximum size of aggregate and the water-cement ratio have been used as the inputs. The compressive strength of concrete has been used as the output in the hybrid ICA-ANN model. Results have been compared with the multiple-linear regression model (MLR), the genetic algorithm (GA) and particle swarm optimization (PSO). The results indicate the superiority and high accuracy of the hybrid ICA-ANN model in predicting the compressive strength of concrete when compared to the other methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.211-213
• /
• 2017

최근 컴퓨터 네트워크를 활용하는 다양한 기기들이 개발되고 급격히 확산되면서, 컴퓨터 네크워크는 전보다 많은 보안문제에 직면하게 되었다. 이에 따라 네트워크 보안을 위한 침입탐지시스템의 필요성이 대두된다. 침입탐지시스템을 구현하기 위한 대표적인 데이터 셋으로는 KDD CUP 99(KDD'99)와 이후 KDD'99의 문제점을 보완하여 공개된 NSL-KDD가 있다. 본 논문에서는 KDD'99와 NSL-KDD를 소개하고 인공신경망을 통해 두 데이터 셋을 비교 분석하였다. Multi-Layer Perceptron을 사용해 데이터 셋을 분석해본 결과, KDD'99는 전체 정확도에서 더 높은 결과를 얻은 반면 공격 별 탐지 정확도 면에서는 NSL-KDD에 뒤쳐졌다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.305-308
• /
• 2007

본 논문은 교통표지판을 자동으로 인식하는 방법에 관한 연구로 기존의 교통표지판 인식시스템에서는 인식하는데 걸리는 시간이 길고 잡음환경에서 인식률이 저하되며 변경된 교통표지판은 인식하지 못하는 문제점이 있다. 본 논문에서는 이와 같은 문제점을 해결하기위해 컬러정보를 이용하여 교통표지판 영역을 추출하고 추출된 이미지를 인식하는데 다층퍼셉트론 신경망 알고리즘을 적용하여 교통표지판 인식시스템을 제안한다. 제안된 방법은 교통표지판의 컬러를 분석하여 영상에서 교통표지판 영역을 추출한다. 영역을 추출하는 방법은 RGB 컬러 공간으로부터 YUV, YIQ, CMYK 컬러 공간이 가지는 특성을 이용한다. 형태처리는 교통표지판의 기하학적 특성을 이용하여 군집화한다. 교통표지판 인식은 학습이 가능한 다층퍼셉트론의 오류역전파알고리즘을 적용하여 인식한다. 다층퍼셉트론 신경망 알고리즘은 패턴인식 분야에서 우수한 성능이 입증 되었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.710-717
• /
• 1993

In this paper, we focus upon the design and applications of adaptive fuzzy-neuro controllers. An intelligent control system is proposed by exploiting the merits of two paradigms, a fuzzy logic controller and a neural network, assuming that we can modify in real time the consequential parts of the rulebase with adaptive learning, and that initial fuzzy control rules are established in a temporarily stable region. We choose the structure of fuzzy hypercubes for the fuzzy controller, and utilize the Perceptron learning rule in order to update the fuzzy control rules on-line with the output error. And, the effectiveness and the robustness of this intelligent controller are shown with application of the proposed adaptive fuzzy-neuro controller to control of the cart-pole system.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.4
• /
• pp.55-69
• /
• 1999

일반적으로 잉크젯 프린터는 농도에 대해 선형적인 계조재현 특성을 나타낸다. 그러나 인간 시각의 경우 농도에 선형적인 프린터 출력에 대하여 비선형적인 지각반응을 나타낸다. 즉 농도가 큰 패치(patch)에 대해서는 명도나 색차에 대한 변별력이 작으며 농도가 작은 패치에 대해서는 좀 더 예민한 변별력을 갖는다. 따라서 농도에 선형적인 프린터 출력은 시각적인 활성영역을 줄이므로 프린터에서 구별되는 계조의 범위가 좁아진다. 그러므로 본 논문에서는 인간의 시지각 특성과 매우 상관도가 높은 CIELAB 색공간을 이용하여 균등한 명도 변화 및 색차를 나타내도록 하는 프린터 계조재현 알고리즘을 제안한다. 이때 시각적으로 균등한 변화를 나타내는 프린터의 입력값을 찾기 위해 다층 퍼셉트론 신경망(multi-layer perceptron neural network, MLP)을 이용하였다. 신경망의 학습을 위해 계조에 따른 패치를 만들고, 프린터 구동입력신호 및 패치의 측정된 값으로 신경망을 학습하였다. 학습된 신경망으로 선형적인 출력을 내는 프린터 구동신호를 찾고 LUT(look-up table)를 이용하여 프린터 입력 신호를 역으로 보정하였다. 결과, 보정된 프린터의 출력이 선형적인 계조 변화를 보였고 변화가 인지되는 계조의 범위가 늘어났으며 실형상에 대한 실험에 있어서도 우수한 화질을 보였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.05a
• /
• pp.509-511
• /
• 2011

본 논문에서는 다층 퍼셉트론 신경회로망과 선형예측계수를 사용한 화자인식 알고리즘을 제안한다. 제안하는 화자인식 알고리즘은 입력받은 음성신호에 대해서 유성음 구간을 추출한다. 추출된 유성음구간에 대하여 선형예측 분석에 의하여 화자의 특성을 가지고 있는 선형예측계수를 구한다. 구해진 선형예측계수를 분류하기 위하여 선형예측계수를 퍼셉트론 신경회로망의 입력으로 사용하여 네트워크의 학습을 수행한다. 본 실험에서는 선형예측계수와 신경회로망을 사용하여 본 화자인식 알고리즘이 유효하다는 것을 인식률을 통하여 확인한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.1
• /
• pp.10-22
• /
• 1993

In this paper, we proposed a Neuro-FAX algorithm having high compression rate and good reconstruction capability in spite of noise and fonts. This algorithm processes the character part and the image part seperately. In the character part, we recognized each characters in document using neural networks, and transmitted the information recognized. And we transmitted the image part as it is by the conventional method. With character set in receiving terminal. it can produce nice document of noise free characters and different font.