• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2010.10a
• /
• pp.108-109
• /
• 2010

본 연구에서는 비선형예측기법으로서 그 우수성을 인정받고 있는 인공신경망모형을 사용하여 컨테이너 물동량 예측을 수행하였다. 그러나 인공신경망모형을 사용해 시계열의 예측결과를 ARIMA모형과 같이 널리 알려진 다른 전통적인 수요예측기법들과 비교 평가한 과거 연구들을 보게 되면 각기 주장하는 바와 그 결론이 상반됨을 알 수 있다. 그래서 인공신경망의 예측성과를 높이기 위한 기존의 선행연구들의 다양한 시도들을 바탕으로 국내 항만의 컨테이너물동량을 예측하고, 그를 통해 여러 모형간의 비교 검증작업을 수행하였다.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2000.04b
• /
• pp.271-276
• /
• 2000

Optimal neuro-control algorithm is extended to the control of a multi-degree-of-freedom structure. An active mass driver(AMD) system on the top roof used as a controller. The control signals are made by a multi-layer perceptron(MLP) which is trained by minimizing a sub-optimal performance index. The performance index is a function of both the output responses and the control signals. Structure having nonlinear hysteretic behavior is also trained and controlled by using proposed control algorithm. Bothe the time delay effect and the dynamics of hydraulic actuator are included in the simulation. Example shows that optimal neuro-control algorithm can be applicable to the multi-degree of freedom structures.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.455-458
• /
• 2001

In this paper, a new license plate recognition method using the morphological size distribution functions and color images is proposed. The proposed method consists of two steps. The first step is license plate extraction process using the plate color and step edge information in the license plate. The second step is the extraction of character feature vectors using the morphological size distribution functions and character recognition process using the MLP(multilayer perceptron). By the use of morphological size distributions functions, the error that may occur during the character region extraction process is lessened and the recognition performances are improved by the decrease of feature vector dimension.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.205-208
• /
• 2003

While multilayer perceptrons (MLPs) have great possibility on the application to speaker verification, they suffer from inferior learning speed. to appeal to users, the speaker verification systems based on MLPs must achieve a reasonable enrolling speed and it is thoroughly dependent on the fast learning of MLPs. To attain real-time enrollment on the systems, the previous two studies have been devoted to the problem and each satisfied the objective. In this paper the two studies are combined md applied to the systems, on the assumption that each method operates on different optimization principle. By conducting experiments using an MLP-based speaker verification system to which the combination is applied on real speech database, the feasibility of the combination is verified from the results of the experiments.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.444-447
• /
• 2005

본 논문에서는, 주성분분석기법과 다층신경망에 기반을 둔 유도전동기의 고장진단기법을 제안하고자 한다. 입력의 수가 많을 경우 다층신경망만을 이용하여 분류하는 데는 한계가 있다. 이러한 문제점을 해결하기 위해 주성분분석기법에 의해 입력특징의 수를 축약한 후, 비선형분류기인 다층신경망을 적용하였다. 또한, 주성분 분석기법에 추출된 특징벡터가 고장상태별로 비선형성특성을 보일 경우 기존의 거리척도 기반에 의한 분류방법으로는 정확한 진단을 하는데 어려움이 있다. 이를 위해 비선형 분류기인 MLP를 적용함으로써 효과적인 고장진단을 하고자 한다. 제안된 기법은 다양한 실험을 통해 기존의 선형분류기에 비해 우수한 결과를 보임을 나타내고자 한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.2
• /
• pp.75-82
• /
• 2000

This paper presents a method of detecting surface flaw of cold-rolled steel plate using image processing technique and a neural network classifier. The amount of steel plate surface image data is reduced by the wavelet transform. Features are extracted from the co-occurence matrix of the partial image corresponding to the low-frequency region, and a MLP neural network classifies into predetermined surface flaw categories. Simulations show the neural network classifier outperforms conventional vector quantization method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.7 no.4
• /
• pp.267-273
• /
• 2007

This paper presents the implementation of an educational kit for intelligent system control education. Neural network control algorithms are presented and control hardware is embedded to control the inverted pendulum system. The RBF network and the MLP network are implemented and embedded on the DSP 2812 chip and other necessary functions are embedded on an FPGA chip. Experimental studies are conducted to compare performances of two neural control methods. The intelligent control educational kit(ICEK) is implemented with the inverted pendulum system whose movements of the cart is limited by space. Experimental results show that the neural controllers can manage to control both the angle and the position of the inverted pendulum systems within a limited distance. Performances of the RCT and the FEL control method are compared as well.

• Proceedings of the KIEE Conference
• /
• 1991.07a
• /
• pp.900-904
• /
• 1991

In this paper, our study for the recognition of Hand-written Korean characters, Arabic numbers and alphabets by neural netwoks. This System extracts feature of character by using the MESH feature point of handwritten character, Arabic numbers and alphabets. To reduce the input image data, features are extracted from each input images. A MLP(multi-layer perceptron) with one hidden layer was trained with a modified BEP(back error propagation) algorithm. This method extracts feature sets of the characters directly from the scanner and can enhance computation speed without using the special preprocesses such as size normalization, smoothing, and thinning.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.10 no.4
• /
• pp.314-318
• /
• 2010

Predicting Alpha-helicies, Beta-sheets and Turns of a proteins secondary structure is a complex non-linear task that has been approached by several techniques such as Neural Networks, Genetic Algorithms, Decision Trees and other statistical or heuristic methods. This project introduces a new machine learning method by combining Bayesian Inference with offline trained Multilayered Perceptron (MLP) models as the likelihood for secondary structure prediction of proteins. With varying window sizes of neighboring amino acid information, the information is extracted and passed back and forth between the Neural Net and the Bayesian Inference process until the posterior probability of the secondary structure converges.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.371-373
• /
• 2003

웹 문서를 분류하는 목적은 특정 주제별로 중요한 문서들을 구분하려는 것과 사용자의 선호도를 바탕으로 개인화를 하려는 것으로 나누어 볼 수 있다. 특히, 웹의 효율적인 탐색을 위해 사용자가 관심 있어 할 웹 문서를 분류하는 것은 중요하다 일반적으로 하나의 웹 문서는 특징 추출방법에 의해 문서 벡터로 표시되며 사용자의 선호여부나 주제번호를 클래스로 삼는다. 사용자가 선호도를 표시한 웹 문서를 사용하여 새로운 웹 문서의 선호 여부를 예측하기 위해 자기 구성지도(SOM)를 사용하면, 시각적으로 구조를 보여주어 데이터 사이의 관계를 효과적으로 이해할 수 있다. 그러나 SOM은 노드의 개수와 구조를 자동적으로 결정하지 못하는 단점이 있기 때문에, SOM의 장점을 활용하면서 자동적으로 구조를 결정하기 위해 구조적응 자기구성지도(SASOM)를 이용한다. 보다 나은 성능과 다양한 해석을 위해, 여러 개의 SASOM을 서로 다른 특징추출 방법을 이용하여 학습시킨 후 사용자가 주관적으로 분류기의 중요도를 결정할 수 있는 퍼지적분을 사용하여 결합하였다. UCI Syskill & Webert 데이터에 대한 실험결과 기존의 DT, MLP, naive Bayes 분류기 보다 향상된 성능을 보였다.