• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1738-1741
• /
• 1997

This paper addresses the issue fo target classification and localization with a SONAR for mobiler robot indoor navigation. In particular, multiple refetions of SONAR sound are used actively and interntionally. As for the SONAR sensor, the multiple reflection has been generally considered as one of the noisy phenomena, which is inevitable in the indoor environments. However, these multiple reflections can be a clue for classifying and localizing targets in the indoor environment if those can be controlled and used well. This paper develops a new SONAR sensor module with a reflection plane which can actively create the multiple refection. This paper also intends to suggest a new target classification emthod which uses the multiple refectiions. We approximate the world as being two dimensional and assume that the targets consisting of the indoor environment are pland, corner, and edge. Multiple reflection paths of an acoustic bean by a SONAR are analyzed, by simulations and the patterns of the TOPs (Time Of Flight) and angles of multiple reflections from each target are also analyzed. In addition, a new algorithm for target classification and localization is proposed.

• Spatial Information Research
• /
• v.12 no.1
• /
• pp.111-125
• /
• 2004

Improvements of classification accuracy are main issues in satellite image classification. Considering the facts that multiple images in the same area are available, there are needs on researches aiming improvements of classification accuracy using multiple data sets. In this study, orthogonal summation method of Dempster-Shafer theory (theory of evidence) is proposed as a multiple imagery classification method and posterior probabilities and classification uncertainty are used in calculation process. Accuracies of the proposed method are higher than conventional classification methods, maximum likelihood classification(MLC) of each data and MLC of merged data sets, which can be certified through statistical tests of mean difference.

• Journal of KIISE:Software and Applications
• /
• v.29 no.3
• /
• pp.192-201
• /
• 2002

Automated text classification is considered as an important method to manage and process a huge amount of documents in digital forms that are widespread and continuously increasing. Recently, text classification has been addressed with machine learning technologies such as k-nearest neighbor, decision tree, support vector machine and neural networks. However, only few investigations in text classification are studied on real problems but on well-organized text corpus, and do not show their usefulness. This paper proposes and analyzes text classification methods for a real application, email document classification task. First, we propose a combining method of multiple neural networks that improves the performance through the combinations with maximum and neural networks. Second, we present another strategy of combining multiple machine learning classifiers. Voting, Borda count and neural networks improve the overall classification performance. Experimental results show the usefulness of the proposed methods for a real application domain, yielding more than 90% precision rates.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.6-10
• /
• 2007

Classification of hyperspectral images is challenging. A very high dimensional input space requires an exponentially large amount of data to adequately and reliably represent the classes in that space. In other words in order to obtain statistically reliable classification results, the number of necessary training samples increases exponentially as the number of spectral bands increases. However, in many situations, acquisition of the large number of training samples for these high-dimensional datasets may not be so easy. This problem can be overcome by using multiple classifiers. In this paper we compared the effectiveness of two approaches for creating multiple classifiers, feature selection and feature extraction. The methods are based on generating multiple feature subsets by running feature selection or feature extraction algorithm several times, each time for discrimination of one of the classes from the rest. A maximum likelihood classifier is applied on each of the obtained feature subsets and finally a combination scheme was used to combine the outputs of individual classifiers. Experimental results show the effectiveness of feature extraction algorithm for generating multiple classifiers.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.852-855
• /
• 2003

This paper reports the RFC(Recursive Flow Classification) algorithm that is available on multiple fields. It is easy to be implemented by both software and hardware. For high speed classification of packets, the implementation of RFC is essential by hardware. Hence, in this paper, RFC algorithm is simulated by Verilog-HDL, and it verify the efficiency of the algorithm. The result shows that the algorithm can perform a packet classification within several cycles. It is not only much faster than software implementation but also enough to support OC192c.

• Journal of Information Processing Systems
• /
• v.15 no.1
• /
• pp.116-126
• /
• 2019

In this paper, an improved one-against-one support vector machine algorithm is used to classify multiple power quality disturbances. To solve the problem of parameter selection, an improved particle swarm optimization algorithm is proposed to optimize the parameters of the support vector machine. By proposing a new inertia weight expression, the particle swarm optimization algorithm can effectively conduct a global search at the outset and effectively search locally later in a study, which improves the overall classification accuracy. The experimental results show that the improved particle swarm optimization method is more accurate than a grid search algorithm optimization and other improved particle swarm optimizations with regard to its classification of multiple power quality disturbances. Furthermore, the number of support vectors is reduced.

• Journal of Korea Multimedia Society
• /
• v.24 no.4
• /
• pp.550-557
• /
• 2021

Image steganalysis is a technique for detecting images with steganographic algorithms applied, called stego images. With state-of-the-art CNN-based steganalysis methods, we can detect stego images with high accuracy, but it is not possible to know which steganographic algorithm is used. Identifying stego images is essential for extracting embedded data. In this paper, as the first step for extracting data from stego images, we propose a hierarchical CNN structure for senary classification of steganographic algorithms. The hierarchical CNN structure consists of multiple CNN networks which are trained to classify each steganographic algorithm and performs binary or ternary classification. Thus, it classifies multiple steganogrphic algorithms hierarchically and stepwise, rather than classifying them at the same time. In experiments of comparing with several conventional methods, including those of classifying multiple steganographic algorithms at the same time, it is verified that using the hierarchical CNN structure can greatly improve the classification accuracy.

• Communications for Statistical Applications and Methods
• /
• v.6 no.2
• /
• pp.383-395
• /
• 1999

This study is concerned with the evaluation of predictive ability of classification models with ordered multiple categories. If categories can be ordered or ranked the spread of misclassification should be considered to evaluate the performance of the classification models using loss rate since the apparent error rate can not measure the spread of misclassification. Since loss rate is known to underestimate the true loss rate the bootstrap method were used to estimate the true loss rate. thus this study suggests the method to evaluate the predictive power of the classification models using loss rate and the bootstrap estimate of the true loss rate.

• Journal of Korean Society for Quality Management
• /
• v.30 no.1
• /
• pp.152-161
• /
• 2002

The analysis of large data sets with hundreds of thousands observations and thousands of independent variables is a formidable computational task. A less parametric method, capable of identifying important independent variables and their interactions, is a tree structured approach to regression and classification. It gives a graphical and often illuminating way of looking at data in classification and regression problems. In this paper, we have reviewed and summarized tile methodology used to construct a tree, multiple trees and the sequential strategy for identifying active compounds in large chemical databases.

• Journal of the Korean Society for Library and Information Science
• /
• v.50 no.2
• /
• pp.101-118
• /
• 2016

For the effective classification of complex subjects, a library classification scheme should adopt multiple division principles (or facets). Each of the multiple principles adopted for the division of complex subjects is sequentially applied at each stage of division. The order of application of these multiple principles during the process of division of complex subjects is called citation order. In order for a classification scheme to be consistent and logical, the citation order of division principles applied to classify complex subjects should be concrete and consistent. Especially, in case of enumerative classification system, decisions on citation order to represent complex subjects significantly affect the structure and organization of the classification system. There are basic principles and theoretical canons of the classification theory on the citation order and its application, but they cannot be applied solidly in the process of classification system development for practical reasons. Therefore, this paper first reviews previous works on classification theories regarding citation order, then explores the conditions and circumstances for the application of citation order.