• Journal of the Korea Society of Computer and Information
• /
• v.14 no.9
• /
• pp.29-35
• /
• 2009

In this paper we propose a mobile ad-hoc routing protocal based on AODV(Ad hoc On demand Distance Vector) with traffic management support and evaluate the performance through simulation. The average reception rate is increased by establishing the shortest route, considering in advance the usable communication resources at each node. For performance evaluation, we analyze the average data reception rate, considering the node mobility.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.577-580
• /
• 2023

물리 기반 시뮬레이션과 같이 연속적인 움직임을 표현하기 위해서 고차 보간(High-order interpolation)을 설계하는 것을 중요한 문제이다. 본 논문에서는 제약적인 벡터와 밀도 형태를 몬테카를로법을 사용하여 이동최소제곱(MLS, Moving least squares)을 제곱하여 이를 통해 속도 필드를 표현할 수 있는 방법을 제안한다. 결과적으로 밀도의 형태를 고려하여 MLS의 가중치가 적용된 결과를 보여주며, 그 결과가 벡터 보간에 얼마나 큰 영향을 끼치는지를 다양한 실험을 통해 보여준다.

• International Journal of Computer Science & Network Security
• /
• v.24 no.3
• /
• pp.101-106
• /
• 2024

This article aims to compare the accuracy of the Backpropagation Neural Network (BPNN) and Learning Vector Quantization (LVQ) approaches in recognizing Sundanese characters. Based on experiments, the level of accuracy that has been obtained by the BPNN technique is 95.23% and the LVQ technique is 66.66%. Meanwhile, the learning time that has been required by the BPNN technique is 2 minutes 45 seconds and then the LVQ method is 17 minutes 22 seconds. The results indicated that the BPNN technique was better than the LVQ technique in recognizing Sundanese characters in accuracy and learning time.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06a
• /
• pp.347-348
• /
• 2008

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2013.07a
• /
• pp.47-48
• /
• 2013

본 논문에서는 이동 카메라 환경에서 이동물체 추적을 위한 영상 분할에 대하여 연구하였다. 입력영상으로 부터 이동물체영역을 분할하기위하여 입력영상에 대하여 윤곽선을 구한 다음 윤곽선 영역에 대하여 BMA을 이용하여 이동벡터를 구한다. 구해진 이동벡터를 같은 특성의 벡터들을 분류하여 이동물체를 분할한다. 제안된 알고리즘이 다중 이동물체의 분할이 가능하였다.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.6
• /
• pp.366-371
• /
• 2014

This paper reports a machine learning approach for image object detection. Object detection and localization in a wild image, such as a STL-10 image dataset, is very difficult to implement using the traditional computer vision method. A convolutional neural network is a good approach for such wild image object detection. This paper presents an object detection application using a convolutional neural network with pretrained feature vector. This is a very simple and well organized hierarchical object abstraction model.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.4 no.3
• /
• pp.271-276
• /
• 2004

This paper is a research on the dynamic signature verification of error rate which are false rejection rate and false acceptance rate, the size of signature verification engine, the size of the characteristic vectors of a signature, the ability to distinguish similar signatures, the processing speed and so on. Also, we present our efficient user interface and performance results.

• East Asian mathematical journal
• /
• v.22 no.1
• /
• pp.1-16
• /
• 2006

Some new reverses of the Schwarz inequality in inner product spaces and applications for vector-valued sequnces and integrals are given.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.9
• /
• pp.125-139
• /
• 2014

In this paper, I propose a response modeling with a Semi-Supervised Support Vector Regression (SS-SVR) algorithm. In order to increase the accuracy and profit of response modeling, unlabeled data in the customer dataset are used with the labeled data during training. The proposed SS-SVR algorithm is designed to be a batch learning to reduce the training complexity. The label distributions of unlabeled data are estimated in order to consider the uncertainty of labeling. Then, multiple training data are generated from the unlabeled data and their estimated label distributions with oversampling to construct the training dataset with the labeled data. Finally, a data selection algorithm, Expected Margin based Pattern Selection (EMPS), is employed to reduce the training complexity. The experimental results conducted on a real-world marketing dataset showed that the proposed response modeling method trained efficiently, and improved the accuracy and the expected profit.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.2
• /
• pp.39-49
• /
• 2000

In this paper, we propose an optimizing feature extraction method for multiclass problems assuming normal distributions. Initially, We start with an arbitrary feature vector Assuming that the feature vector is used for classification, we compute the classification error Then we move the feature vector slightly in the direction so that classification error decreases most rapidly This can be done by taking gradient We propose two search methods, sequential search and global search In the sequential search, an additional feature vector is selected so that it provides the best accuracy along with the already chosen feature vectors In the global search, we are not constrained to use the chosen feature vectors Experimental results show that the proposed algorithm provides a favorable performance.