• The Journal of the Acoustical Society of Korea
• /
• v.10 no.4
• /
• pp.38-47
• /
• 1991

The frequency variation among speakers is one of problems in the speech recognition. This paper applies fuzzy theory to solve the variation problem of frequency features. Reference patterns are expressed by fuzzified patterns which are produced by the peak frequency and the peak energy extracted from codebooks which are generated from training words uttered by several speakers, as they should include common features of speech signals. Words are recognized by fuzzy inference which uses the certainty factor between the reference patterns and the test fuzzified patterns which are produced by the peak frequency and the peak energy extracted from the power spectrum of input speech signals. Practically, in computing the certainty factor, to reduce memory capacity and computation requirements we propose a new equation which calculates the improved certainty factor using only the difference between two fuzzy values. As a result of experiments to test this word recognition method by fuzzy interence with Korean digits, it is shown that this word recognition method using the new equation presented in this paper, can solve the variation problem of frequency features and that the memory capacity and computation requirements are reduced.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.6
• /
• pp.58-67
• /
• 2009

A correspondence matching is one of the important tasks in computer vision, and it is not easy to find corresponding points in variable environment where a scale, rotation, view point and illumination are changed. A SURF(Speeded Up Robust Features) algorithm have been widely used to solve the problem of the correspondence matching because it is faster than SIFT(Scale Invariant Feature Transform) with closely maintaining the matching performance. However, because SURF considers only gray image and local geometric information, it is difficult to match corresponding points on the image where similar local patterns are scattered. In order to solve this problem, this paper proposes an extended SURF algorithm that uses the invariant color and global geometric information. The proposed algorithm can improves the matching performance since the color information and global geometric information is used to discriminate similar patterns. In this paper, the superiority of the proposed algorithm is proved by experiments that it is compared with conventional methods on the image where an illumination and a view point are changed and similar patterns exist.

• Journal of Information Technology Services
• /
• v.18 no.2
• /
• pp.143-159
• /
• 2019

Machine learning (ML) is a method of fitting given data to a mathematical model to derive insights or to predict. In the age of big data, where the amount of available data increases exponentially due to the development of information technology and smart devices, ML shows high prediction performance due to pattern detection without bias. The feature engineering that generates the features that can explain the problem to be solved in the ML process has a great influence on the performance and its importance is continuously emphasized. Despite this importance, however, it is still considered a difficult task as it requires a thorough understanding of the domain characteristics as well as an understanding of source data and the iterative procedure. Therefore, we propose methods to apply deep learning for solving the complexity and difficulty of feature extraction and improving the performance of ML model. Unlike other techniques, the most common reason for the superior performance of deep learning techniques in complex unstructured data processing is that it is possible to extract features from the source data itself. In order to apply these advantages to the business problems, we propose deep learning based methods that can automatically extract features from transaction data or directly predict and classify target variables. In particular, we applied techniques that show high performance in existing text processing based on the structural similarity between transaction data and text data. And we also verified the suitability of each method according to the characteristics of transaction data. Through our study, it is possible not only to search for the possibility of automated feature extraction but also to obtain a benchmark model that shows a certain level of performance before performing the feature extraction task by a human. In addition, it is expected that it will be able to provide guidelines for choosing a suitable deep learning model based on the business problem and the data characteristics.

• Nuclear Engineering and Technology
• /
• v.55 no.6
• /
• pp.2096-2106
• /
• 2023

Rotating machinery is widely applied in important equipment of nuclear power plants (NPPs), such as pumps and valves. The research on intelligent fault diagnosis of rotating machinery is crucial to ensure the safe operation of related equipment in NPPs. However, in practical applications, data-driven fault diagnosis faces the problem of small and imbalanced samples, resulting in low model training efficiency and poor generalization performance. Therefore, a deep convolutional conditional generative adversarial network (DCCGAN) is constructed to mitigate the impact of imbalanced samples on fault diagnosis. First, a conditional generative adversarial model is designed based on convolutional neural networks to effectively augment imbalanced samples. The original sample features can be effectively extracted by the model based on conditional generative adversarial strategy and appropriate number of filters. In addition, high-quality generated samples are ensured through the visualization of model training process and samples features. Then, a deep convolutional neural network (DCNN) is designed to extract features of mixed samples and implement intelligent fault diagnosis. Finally, based on multi-fault experimental data of motor and bearing, the performance of DCCGAN model for data augmentation and intelligent fault diagnosis is verified. The proposed method effectively alleviates the problem of imbalanced samples, and shows its application value in intelligent fault diagnosis of actual NPPs.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.9
• /
• pp.9-23
• /
• 2024

In this study, we propose a method to improve the classification accuracy of body mass index (BMI) estimation techniques based on three-dimensional gait data. In previous studies on BMI estimation techniques, the classification accuracy was only about 60%. In this study, we identify the reasons for the low BMI classification accuracy. According to our analysis, the reason is the use of the undersampling technique to address the class imbalance problem in the gait dataset. We propose applying oversampling instead of undersampling to solve the class imbalance issue. We also demonstrate the usefulness of anthropometric and spatiotemporal features in gait data-based BMI estimation techniques. Previous studies evaluated the usefulness of anthropometric and spatiotemporal features in the presence of undersampling techniques and reported that their combined use leads to lower BMI estimation performance than when using either feature alone. However, our results show that using both features together and applying an oversampling technique achieves state-of-the-art performance with 92.92% accuracy in the BMI estimation problem.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1309-1314
• /
• 2003

If an agent has a learning ability using previous knowledge, then it is expected that the agent can speed up learning by interacting with environment. In this paper, we present an improved reinforcement learning algorithm using domain knowledge which can be represented by problem-independent features and their classifiers. Here, neural networks are employed as knowledge classifiers. To show the validity of our proposed algorithm, computer simulations are illustrated, where navigation problem of a mobile robot and a micro aerial vehicle(MAV) are considered.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.7
• /
• pp.78-88
• /
• 1997

A three dimensional-two spatical dimensions plus time-image segmentation is widely used in a very low bit rate image sequence coding because it can solve the region correspondence problem. Mathematical morphology is a very efficient tool for the segmentation because it deals well with geometric features such as size, shape, contrast and connectivity. But if the motion in the image sequence is large in time axis, the conventional 3D morphological segmentation algorithm have difficulty in solving region correspondence problem. To alleviate this problem, we propose the hierarchical image sequence segmentation algorithm that uses the region motion information. Since the motion of a region in previous level affects that in current level uses the previous motion information to increase region correspondence. Simulation result shows improved performance for sequence frames with large motion.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.3
• /
• pp.614-620
• /
• 2009

The force distribution in multi-legged robots is a constrained, optimization problem. The solution to the problem is the set points of the leg contact forces for a particular system task. In this paper, an efficient and general formulation of the force distribution problem is developed using linear programming. The considered walking robot is a quadruped robot with a locked-joint failure, i.e., a joint of the failed leg is locked at a known place. For overcoming the drawback of marginal stability in fault-tolerant gaits, we define safety margin on friction constraints as the objective function to be maximized. Dynamic features of locked-joint failure are represented by equality and inequality constraints of linear programming. Unlike the former study, our result can be applied to various forms of walking such as crab and turning gaits. Simulation results show the validity of the proposed scheme.

• Korean Management Science Review
• /
• v.10 no.2
• /
• pp.11-26
• /
• 1993

This paper presents a new interactive procedure for multiple objective linear programming problem (MOLP). In practical multiple objective linear programming applications, there is usually no need for the decision maker to consider solutions which are not efficient. Therefore, the interactive procedure presented here searches only among efficient solutions and terminates with a solution that is guaranteed to be efficient. It also can converge to nonextreme efficient final solutions rather than being restricted to only extreme efficient points of the feasible set. The procedure does not require sophisticated judgements or inputs from the decision maker. One of the most attractive features of the procedure however, is that the method allows the DM to examine the efficient faces it finds. As iteration goes, the DM can explore a wide variety of efficient faces rather than efficient faces confined to only certain subregion of the feasible set of problem MOLP since the efficient faces that the procedure finds need not be adjacent. This helps the DM explore the nature of the efficient set of problem MOLP and also helps the DM have confidence with a final solution. For these reasons, I feel that the procedure offer significant promise in solving multiple objective linear programs rapidly and in a satisfying manner to the DM.

• Structural Engineering and Mechanics
• /
• v.3 no.4
• /
• pp.359-372
• /
• 1995

This study presents an efficient method for optimum design of plate and shell structures, when the design variables are continuous or discrete. Both sizing and shape design variables are considered. First the structural responses such as element forces are approximated in terms of some intermediate variables. By substituting these approximate relations into the original design problem, an explicit nonlinear approximate design task with high quality approximation is achieved. This problem with continuous variables, can be solved by means of numerical optimization techniques very efficiently, the results of which are then used for discrete variable optimization. Now, the approximate problem is converted into a sequence of second level approximation problems of separable form and each of which is solved by a dual strategy with discrete design variables. The approach is efficient in terms of the number of required structural analyses, as well as the overall computational cost of optimization. Examples are offered and compared with other methods to demonstrate the features of the proposed method.