• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.6
• /
• pp.238-245
• /
• 2013

Gesture recognition has been widely one of the research areas for natural user interface. This paper presents a novel gesture recognition method using tree classification and multiclass SVM(Support Vector Machine). In the learning step, 3D trajectory of human gesture obtained by a Kinect sensor is classified into the tree nodes according to their distributions. The gestures are resampled and we obtain the histogram of the chain code from the normalized data. Then multiclass SVM is applied to the classified gestures in the node. The input gesture classified using the constructed tree is recognized with multiclass SVM.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.746-756
• /
• 1995

This paper describes the intractive visual programming environment using multi-view which shows the tools of visualization for called and the visualizations for called member-function flow in C＋＋ language. This research defines new visual symbols for class and constructs interactive visual programming environment of various views by using visual symbols. Our proposed interactive multi-view visual programming environment can represent visualization for representation of class and execution relationships between objects in the object-oriented language, which is easy to understand the structure of object-oriented program, therefore our proposed interactive visual programming environment enables easy program development, and can use of education and trainning for beginner in useful.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.9 no.5
• /
• pp.472-479
• /
• 1999

In this paper, we propose a land cover pattern classifier for remote sensing image by using neuro-fuzzy algorithm. The proposed pattem classifier has a 3-layer feed-forward architecture that is derived from generic fuzzy perceptrons, and the weights are con~posed of h u y sets. We also implement a neuro-fuzzy pattern classification system in the Visual C++ environment. To measure the performance of this, we compare it with the conventional neural networks with back-propagation learning and the Maximum-likelihood algorithms. We classified the remote sensing image into the eight classes covered the majority of land cover feature, selected the same training sites. Experimental results show that the proposed classifier performs well especially in the mixed composition area having many classes rather than the conventional systems.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.681-687
• /
• 2001

In order to raise a class discrimination power by combining multiple classifiers under the Bayesian decision theory, the upper bound of a Bayes error rate bounded by the conditional entropy of a class variable and decision variables obtained from training data samples should be minimized. Wang and Wong proposed a tree dependence first-order approximation scheme of a high order probability distribution composed of the class and multiple feature pattern variables for minimizing the upper bound of the Bayes error rate. This paper presents an extended high order product approximation scheme dealing with higher order dependency more than the first-order tree dependence, based on the minimization of the upper bound of the Bayes error rate. Multiple recognizers for unconstrained handwritten numerals from CENPARMI were combined by the proposed approximation scheme using the Bayesian formalism, and the high recognition rates were obtained by them.

• The Journal of the Korea Contents Association
• /
• v.9 no.1
• /
• pp.446-454
• /
• 2009

In this paper, we propose u-learning hub site systems which are designed and implemented based on learning activities oriented components. The proposed systems are composed of component which can process the functionalities for coming into action of learning activities through various devices. Specially, each component is broken into class units by which learning activities of users can be performed on various devices. When users try to connect u-learning system in hub site, the system explores devices of users and connection program and then selects components that are fit to the activities and combines them in realtime. Through the methodology of u-learning hub site, the system proposed in this paper provides u-learning environment so that users can use the learning activity services taking no influence on time, place, devices and programs under the consistent system. That is different to traditional e-learning system which cannot support various devices of users directly.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.885-896
• /
• 2022

Motor failure in manufacturing plays an important role in future A/S and reliability. Motor failure is detected by measuring sound, current, and vibration. For the data used in this paper, the sound of the car's side mirror motor gear box was used. Motor sound consists of three classes. Sound data is input to the network model through a conversion process through MelSpectrogram. In this paper, various methods were applied, such as data augmentation to improve the performance of classifying fault motors and various methods according to class imbalance were applied resampling, reweighting adjustment, change of loss function and representation learning and classification into two stages. In addition, the curriculum learning method and self-space learning method were compared through a total of five network models such as Bidirectional LSTM Attention, Convolutional Recurrent Neural Network, Multi-Head Attention, Bidirectional Temporal Convolution Network, and Convolution Neural Network, and the optimal configuration was found for motor sound classification.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10b
• /
• pp.416-418
• /
• 1998

이기종간의 실행 환경에 적합하도록 설계된 자바 바이트코드는 각 플랫폼에서 인터프리터 방식으로 수행되므로 효율성에서 단점을 가지고 있다. 따라서 바이트코드의 효율적인 실행을 위해 목적 코드로의 변환 과정이 요구된다. 목적 코드로 변환 과정에서 자바의 실행 환경 또한 목적 기계의 실행 환경으로 이전이 이루어져야 한다. 본 논문에서는 자바 실행 환경을 SPARC 실행 환경으로 이전하는 방법을 제시한다. 먼저, 자바 가상 기계의 구성 요소를 분석한 후 이에 대응하는 SPARC 실행 환경에 필요한 각각의 구성 요소들을 제시하고, SPARC 실행 환경에 맞는 모델을 구축한다. 마지막으로, 자바에서 제공하는 표준 클래스 라이브러리에 대해서는 SPARC 라이브러리 형태로 변환하여 제공하게 된다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.85-87
• /
• 2002

본 논문은 씬 서버들을 클러스터로 구성한 서비스 서버들의 모니터링 항목 값을 인터넷으로 연결된 모니터링 서버에서 수집할 수 있는 모니터링 시스템을 구성한다. 본 시스템은 MS .NET 플랫폼에서 제공하는 성능 모니터 클래스를 사용하여 구현하였으며, 일반적인 모니터링 시스템과 달리 토픽기반 메시지 처리기를 사용하여 다양한 클라이언트의 요구에 적절히 대응할 수 있고, 지역 네트워크뿐만 아니라 인터넷으로 연결된 원격 네트워크의 클라이언트로 수집된 모니터링 항목 값을 전송할 수 있다.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1746-1747
• /
• 2007

본 논문에서는 다층 퍼셉트론(Multi-Layer Perceptron)을 이용한 하드 디스크 결함 분포의 패턴 인식 기법을 제시한다. 결함 분포로부터 5 가지의 특징들을 추출하고, 이를 이용하여 퍼셉트론의 입력을 구성하였으며, 미리 분류된 표준 패턴 클래스를 이용하여 퍼셉트론의 출력을 구성하였다. 테스트 결과, 제시된 신경망은 하드 디스크의 패턴 분류에 만족할 만한 성능을 나타내었다.

• Journal of Korea Multimedia Society
• /
• v.1 no.2
• /
• pp.239-248
• /
• 1998

Software reengineering is making various research for solutions against problem of maintain existing system. Reengineering has a meaning of development of softwares on existing systems through the reverse-engineering and the forward-engineering. It extracts classes from existing system's softwares to increase the comprehension of the system and enhance the maintenability of softwares. Most of the important concepts used in reengineering is composition that is restructuring of the existing objects from other components. The classes and clusters in storage have structural relationship with system's main components to reuse in the higher level. These are referenced as dynamic informations through structuring an architect for each of them. The classes are created by extractor, searcher and composer through representing existing object-oriented source code. Each of classes and clusters extract refined informations through optimization. New architecture is created from the cluster based on its classes' relationship in storage. This information can be used as an executable code later on. In this paper, we propose the tools, it presented by this thesis presents a new information to users through analysing, based on reengineering, Object-Oriented informations and practicing composition methodology. These composite classes will increase reusability and produce higher comprehension information to consist maintainability for existing codes.