• The Transactions of the Korea Information Processing Society
• /
• v.6 no.1
• /
• pp.253-260
• /
• 1999

In this paper, we suggest a new direct method for mage segmentation using texture and color information combined through a multivariate linear discriminant algorithm. The color texture is computed in nin 3${\times}$3 masks obtained from each 3${\times}$3${\times}$3 spatio-spectral neighborhood in the image using the classical haralick and Pressman texture features. Among these 9${\times}$28 texture features the best set was extracted from a training set. The resulting set of 10 features were used to segment an image into four different regions. The resulting segmentation was Compared to classical color and texture segmentation methods using both box classifiers and maximum likelihood classification. It compared favourably on the test image from a Fastred-Lightgreen stained prostatic histological tissue section based on visual inspection. The classification accuracy of 97.5% for the new method obtained on the training data was also among the best of the tested methods. If these results hold for a larger set of images, this method should be a useful tool for segmenting images where both color and texture are relevant for the segmentation process.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.3033-3044
• /
• 1997

This paper presents a system which recognizes dynamic hand gesture for virtual reality (VR). A dynamic hand gesture is a method of communication for human and computer who uses gestures, especially both hands and fingers. Since the human hands and fingers are not the same in physical dimension, the produced by two persons with their hands may not have the same numerical values where obtained through electronic sensors. To recognize meaningful gesture from continuous gestures which have no token of beginning and end, this system segments current motion states using the state automata. In this paper, we apply a fuzzy min-max neural network and feature analysis method using fuzzy logic for on-line pattern recognition.

• The KIPS Transactions:PartB
• /
• v.14B no.1 s.111
• /
• pp.59-64
• /
• 2007

In these days, malicious codes have become reality and evolved significantly to become one of the greatest threats to the modern society where important information is stored, processed, and accessed through the internet and the computers. Computer virus is a common type of malicious codes. The standard techniques in anti-virus industry is still based on signatures matching. The detection mechanism searches for a signature pattern that identifies a particular virus or stain of viruses. Though more accurate in detecting known viruses, the technique falls short for detecting new or unknown viruses for which no identifying patterns present. To cope with this problem, anti-virus software has to incorporate the learning mechanism and heuristic. In this paper, we propose a fuzzy diagnosis system(FDS) using fuzzy c-means algorithm(FCM) for the cluster analysis and a decision status measure for giving a diagnosis. We compare proposed system FDS to three well known classifiers-KNN, RF, SVM. Experimental results show that the proposed approach can detect unknown viruses effectively.

• The KIPS Transactions:PartB
• /
• v.19B no.1
• /
• pp.19-26
• /
• 2012

The rapid increase of information imposes new demands of content management. The purpose of automatic audio segmentation and classification is to meet the rising need for efficient content management. With this reason, this paper proposes a high-accuracy algorithm that segments audio signals and classifies them into different classes such as speech, music, silence, and environment sounds. The proposed algorithm utilizes support vector machine (SVM) to detect audio-cuts, which are boundaries between different kinds of sounds using the parameter sequence. We then extract feature vectors that are composed of statistical data and they are used as an input of fuzzy c-means (FCM) classifier to partition audio-segments into different classes. To evaluate segmentation and classification performance of the proposed SVM-FCM based algorithm, we consider precision and recall rates for segmentation and classification accuracy for classification. Furthermore, we compare the proposed algorithm with other methods including binary and FCM classifiers in terms of segmentation performance. Experimental results show that the proposed algorithm outperforms other methods in both precision and recall rates.

• The Korean Journal of Applied Statistics
• /
• v.18 no.2
• /
• pp.343-354
• /
• 2005

The goal of this paper is to compare classification performances and to find a better classifier based on the characteristics of data. The compared methods are CART with two ensemble algorithms, bagging or boosting and SVM. In the empirical study of twenty-eight data sets, we found that SVM has smaller error rate than the other methods in most of data sets. When comparing bagging, boosting and SVM based on the characteristics of data, SVM algorithm is suitable to the data with small numbers of observation and no missing values. On the other hand, boosting algorithm is suitable to the data with number of observation and bagging algorithm is suitable to the data with missing values.

• Korean Journal of Remote Sensing
• /
• v.30 no.2
• /
• pp.207-218
• /
• 2014

The information on the land cover along stream corridor is important for stream restoration and maintenance activities. This study aims to review the different classification methods for mapping the status of stream corridors in Seom River using airborne RGB and CIR digital ortho imagery with a ground pixel resolution of 0.2m. The maximum likelihood classification, minimum distance classification, parallelepiped classification, mahalanobis distance classification algorithms were performed with regard to the improvement methods, the skewed data for training classifiers and filtering technique. From these results follows that, in aerial image classification, Maximum likelihood classification gave results the highest classification accuracy and the CIR image showed comparatively high precision.

• The Korean Journal of Applied Statistics
• /
• v.18 no.2
• /
• pp.421-434
• /
• 2005

Machine learning has been extensively studied in recent years as effective tools in pattern classification problem. Although there have been several approaches to machine learning, we focus on the mathematical programming (in particular, multi-objective and goal programming; MOP/GP) approaches in this paper. Among them, Support Vector Machine (SVM) is gaining much popularity recently. In pattern classification problem with two class sets, the idea is to find a maximal margin separating hyperplane which gives the greatest separation between the classes in a high dimensional feature space. However, the idea of maximal margin separation is not quite new: in 1960's the multi-surface method (MSM) was suggested by Mangasarian. In 1980's, linear classifiers using goal programming were developed extensively. This paper proposes a new family of SVM using MOP/GP techniques, and discusses its effectiveness throughout several numerical experiments.

• Journal of the Korean Wood Science and Technology
• /
• v.45 no.6
• /
• pp.797-808
• /
• 2017

Automatic wood species identification systems have enabled fast and accurate identification of wood species outside of specialized laboratories with well-trained experts on wood species identification. Conventional automatic wood species identification systems consist of two major parts: a feature extractor and a classifier. Feature extractors require hand-engineering to obtain optimal features to quantify the content of an image. A Convolutional Neural Network (CNN), which is one of the Deep Learning methods, trained for wood species can extract intrinsic feature representations and classify them correctly. It usually outperforms classifiers built on top of extracted features with a hand-tuning process. We developed an automatic wood species identification system utilizing CNN models such as LeNet, MiniVGGNet, and their variants. A smartphone camera was used for obtaining macroscopic images of rough sawn surfaces from cross sections of woods. Five Korean softwood species (cedar, cypress, Korean pine, Korean red pine, and larch) were under classification by the CNN models. The highest and most stable CNN model was LeNet3 that is two additional layers added to the original LeNet architecture. The accuracy of species identification by LeNet3 architecture for the five Korean softwood species was 99.3%. The result showed the automatic wood species identification system is sufficiently fast and accurate as well as small to be deployed to a mobile device such as a smartphone.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.7
• /
• pp.37-45
• /
• 2013

Context dependent fusion (CDF) is a fusion algorithm that combines multiple outputs from different classifiers to achieve better performance. CDF tries to divide the problem context into several homogeneous sub-contexts and to fuse data locally with respect to each sub-context. CDF showed better performance than existing methods, however, it is sensitive to noise due to the large number of parameters optimized and the innate linearity limits the application of CDF. In this paper, a variant of CDF using support vector machines (SVMs) for fusion and kernel principal component analysis (K-PCA) for context extraction is proposed to solve the problems in CDF, named CDF-SVM. Kernel PCA can shape irregular clusters including elliptical ones through the non-linear kernel transformation and SVM can draw a non-linear decision boundary. Regularization terms is also included in the objective function of CDF-SVM to mitigate the noise sensitivity in CDF. CDF-SVM showed better performance than CDF and its variants, which is demonstrated through the experiments with a landmine data set.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.73-81
• /
• 2015

The paper proposes a prototype selection method and evaluates the generalization performance of standard algorithms and prototype based classification learning. The proposed prototype classifier defines multidimensional spheres with variable radii within class areas and generates a small set of training data. The nearest-neighbor classifier uses the new training set for predicting the class of test data. By decomposing bias and variance of the mean expected error value, we compare the generalization errors of k-nearest neighbor, Bayesian classifier, prototype selection using fixed radius and the proposed prototype selection method. In experiments, the bias-variance changing trends of the proposed prototype classifier are similar to those of nearest neighbor classifiers with all training data and the prototype selection rates are under 27.0% on average.