• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.360-366
• /
• 2003

Although multilayer perceptrons (MLPs) present several advantages against other pattern recognition methods, MLP-based speaker verification systems suffer from slow enrollment speed caused by many background speakers to achieve a low verification error. To solve this problem, the quantitative discriminative cohort speakers (QnDCS) method, by introducing the cohort speakers method into the systems, reduced the number of background speakers required to enroll speakers. Although the QnDCS achieved the goal to some extent, the improvement rate for the enrolling speed was still unsatisfactory. To improve the enrolling speed, this paper proposes the qualitative DCS (QlDCS) by introducing a qualitative criterion to select less background speakers. An experiment for both methods is conducted to use the speaker verification system based on MLPs and continuants, and speech database. The results of the experiment show that the proposed QlDCS method enrolls speakers in two times shorter time than the QnDCS does over the online error backpropagation(EBP) method.

• Korean Journal of Cognitive Science
• /
• v.10 no.1
• /
• pp.13-24
• /
• 1999

In this paper, we propose a classifier based on modular networks using an unsupervised learning method. The structure of each module is designed through stochastic analysis of input data and each module classifier data independently. The result of independent classification of each module and a measure of the nearest distance are integrated during the final data classification phase to allow more precise c classification. Computation time is decreased by deleting modules that have been classified to be incorrect during the final classification phase. Using this method. a neural network sharing the best performance was implemented without considering. lots of of variables which can affect the performance of the neural network.

• Journal of Biomedical Engineering Research
• /
• v.19 no.1
• /
• pp.1-8
• /
• 1998

Clustered microcalcifications on X-ray mammograms are an important sign for early detection of breast cancer. This paper proposes a computer-aided diagnosis method for the detection of clustered microcalcifications and marking their locations on digitized mammograms. The proposed detection method consists of the region of interest (ROI) selection, the film-artifact removal, the surrounding texture analysis method for the detection of clustered microcalcifications, which is based on the second-order histogram in two nested surrounding regions on the current pixel. This paper also describes the effectiveness of the proposed film-artifact removal filter in terms of the classification performance with the receiver operating-characteristics(ROC) analysis. A three-layer backpropagation neural network is employed as a classifier. The appropriate marking for the locations of clustered microcalcifications can be used to alert radiologists to locations of suspicious lesions.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.4
• /
• pp.767-778
• /
• 2003

This paper presents a method for generating optimal fuzzy rules using a genetic algorithm. Fuzzy rules are generated from the training data in the first stage. In this stage, fuzzy c-Means clustering method and cluster validity are used to determine the structure and initial parameters of the fuzzy inference system. A cluster validity is used to determine the number of clusters, which can be the number of fuzzy rules. Once the structure is figured out in the first stage, parameters relating the fuzzy rules are optimized in the second stage. Weights and variance parameters are tuned using genetic algorithms. Variance parameters are also managed with left and right for asymmetrical Gaussian membership function. The method ensures convergence toward a global minimum by using genetic algorithms in weight and variance spaces.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.2
• /
• pp.80-89
• /
• 2014

In this paper, we conceive a context-aware recommendations system for water distribution in future smart water grids, with taking the end users' profiles, water types, network conditions into account. A spectral clustering approach is developed to cluster end users into different communities, based on the end users' common interests in water resources. A back-propagation (BP) neural network is designed to obtain the rating list of the end users' preferences on water resources and the water resource with the highest prediction rating is recommended to the end users. Simulation results demonstrate that the proposed scheme achieves the improved accuracy of recommendation within 2.5% errors notably together with a better user experience in contrast to traditional recommendations approaches.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.3C
• /
• pp.374-386
• /
• 2004

The mammogram provides the way to observe detailed internal organization of breasts to radiologists for the early detection. This paper is mainly focused on efficiently detecting the Microcalcification's Region Of Interest(ROI)s. Breast cancers can be caused from either microcalcifications or masses. Microcalcifications are appeared in a digital mammogram as tiny dots that have a little higher gray levels than their surrounding pixels. We can roughly determine the area which possibly contain microcalifications. In general, it is very challenging to find all the microcalcifications in a digital mammogram, because they are similar to some tissue parts of a breast. To efficiently detect microcalcifications ROI, we used four sequential processes; preprocessing for breast area detection, modified multilevel thresholding, ROI selection using simple thresholding filters and final ROI selection with two stages of neural networks. The filtering process with boundary conditions removes easily-distinguishable tissues while keeping all microcalcifications so that it cleans the thresholded mammogram images and speeds up the later processing by the average of 86％. The first neural network shows the average of 96.66％ recognition rate. The second neural network performs better by showing the average recognition rate 98.26％. By removing all tissues while keeping microcalcifications as much as possible, the next parts of a CAD system for detecting breast cancers can become much simpler.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.4
• /
• pp.31-38
• /
• 2009

Lens distortion is inevitable phenomenon in machine vision system. More and more distortion phenomenon is occurring in order to choice of lens for minimizing cost and system size. As shown above, correction of lens distortion is critical issue. However previous lens correction methods using camera model have problem such as nonlinear property and complicated operation. And recent lens correction methods using neural network also have accuracy and efficiency problem. In this study, I propose new algorithms for correction of lens distortion. Distorted image is divided based on the distortion quantity using k-means. And each divided image region is corrected by using neural network. As a result, the proposed algorithms have better accuracy than previous methods without image division.

• The Korean Journal of Applied Statistics
• /
• v.15 no.2
• /
• pp.251-267
• /
• 2002

Self Organizing Maps(SOM) by Kohonen is very fast algorithm in neural networks. But it doesn't show sure rules of training results. In this paper, we introduce to Bayesian Learning for Self Organizing Maps(BLSOM) which combines self organizing maps with Bayesian learning. So it supports explanatory power of models and improves prediction. BLSOM has global optima anywhere but SOM has not. This is proved by experiment in this paper.

• Journal of Korea Water Resources Association
• /
• v.34 no.4
• /
• pp.303-316
• /
• 2001

In this study, Radial Basis Function(RBF) Neural Networks Model, a kind of Hybrid Neural Networks was applied to hydrological forecasting in a small watershed. RBF Neural Networks Model has four kinds of parameters in it and consists of unsupervised and supervised training patterns. And Gaussian Kernel Function(GKF) was used among many kinds of Radial Basis Functions(RBFs). K-Means clustering algorithm was applied to optimize centers and widths which ate the parameters of GKF. The parameters of RBF Neural Networks Model such as centers, widths weights and biases were determined by the training procedures of RBF Neural Networks Model. And, with these parameters the validation procedures of RBF Neural Networks Model were carried out. RBF Neural Networks Model was applied to Wi-Stream basin which is one of the IHP Representative basins in South Korea. 10 rainfall events were selected for training and validation of RBF Neural Networks Model. The results of RBF Neural Networks Model were compared with those of Elman Neural Networks(ENN) Model. ENN Model is composed of One Step Secant BackPropagation(OSSBP) and Resilient BackPropagation(RBP) algorithms. RBF Neural Networks shows better results than ENN Model. RBF Neural Networks Model spent less time for the training of model and can be easily used by the hydrologists with little background knowledge of RBF Neural Networks Model.

• Journal of Korea Society of Industrial Information Systems
• /
• v.11 no.1
• /
• pp.77-85
• /
• 2006

In this paper, we propose a hierarchical classifier of object images using neural networks for content-based image classification. The images for classification are object images that can be divided into foreground and background. In the preprocessing step, we extract the object region and shape-based texture features extracted from wavelet transformed images. We group the image classes into clusters which have similar texture features using Principal Component Analysis(PCA) and K-means. The hierarchical classifier has five layes which combine the clusters. The hierarchical classifier consists of 59 neural network classifiers learned with the back propagation algorithm. Among the various texture features, the diagonal moment was the most effective. A test with 1000 training data and 1000 test data composed of 10 images from each of 100 classes shows classification rates of 81.5% and 75.1% correct, respectively.