• Proceedings of the Korea Contents Association Conference
• /
• 2009.05a
• /
• pp.668-671
• /
• 2009

콘텐츠의 네트워크를 통한 배포가 늘어나면서 콘텐츠의 보안에도 많은 연구가 활발히 진행되고 있다. 이러한 콘텐츠 보안은 콘텐츠 자체에 보안을 설정하거나 콘텐츠를 볼 수 있는 프로그램 등에 보안을 설정하는 방법 등을 사용하고 있다. 여러 가지 보안 기술 중 워터마크는 불법 콘텐츠 유포의 방지, 불법 콘텐츠의 추적, 저작권 주장, 위 변조 방지 등에 사용되고 있다. 워터마크는 위 변조를 방지하는 목적에서는 잘 훼손되도록 제작되지만 저작권 보호의 목적에서는 훼손되면 안 된다. 그러나 콘텐츠의 유통 과정 중 워터마크는 훼손되기 쉽다. 콘텐츠를 제작하는 과정에서 워터마크가 삽입되면 콘텐츠의 압축, 편집, 송출 등의 단계에서 공격을 받아 훼손될 수 있기 때문이다. 이러한 공격에서 워터마크를 살아남게 하기 위해서 본 논문에서는 움직임 예측을 이용한 워터마크 삽입 방법을 제안한다. 방법은 워터마크 메시지를 삽입할 때 움직임이 있다면 그 부분에 좀 더 강하게 워터마크 메시지를 삽입하도록 하는 것이다. 실험 결과 이러한 방법으로 워터마크를 삽입할 경우 워터마크가 좀 더 잘 검출됨을 확인할 수 있었다. 제안하는 내용은 콘텐츠 보안이 필요한 VOD, IP-TV, DMB, FTA(Free-to-Air) 방송 등에 사용될 수 있을 것이다.

• Proceedings of the KIPE Conference
• /
• 2004.07b
• /
• pp.635-639
• /
• 2004

This paper presents a simple PWM method to control three phase four leg multilevel inverters, which have been developed for supplying electrical power to three phase unbalanced load and for related power active filters. The method is derived from a general correlation between space vector PWM (SVPWM) method and carrier based PWM (CPWM) method. As an advantage, the simplicity and flexibility of the proposed CPWM control can be obtained and the complicated calculations of the 3-D SVPWM concepts can be avoided. The method has been mathematical formulated and demonstrated by simulation results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06d
• /
• pp.485-490
• /
• 2008

무선 인터넷의 구조적 특성상 한 셀에서 대역폭을 공유하고 그 안에서 각기 다른 QoS를 요구하는 서비스들이 한정된 자원을 사용한다. 트래픽의 변화와 패턴을 예측하기 위한 분석은 실제 서비스를 제공하기 전인 기획단계에서 매우 중요한 도구로 사용이 된다. 무선망의 트래픽을 예측하기 위해서는 유선망의 분석과는 다른 방법이 필요하기 때문에 정확한 분류를 위해서 본 연구에서는 세션의 단위로 분석할 것을 제안한다. 또한 Classification and Regression Tree(CART) 와 Support Vector Machine(SVM) 의 두 개의 판별 분류 기법을 서로 비교하고 그 성능을 평가한다. 두 개의 판별 기법의 오차는 CART의 경우 0.0094 그리고 SVM의 경우 0.0089로 둘 다 우수한 성능을 보였지만 쉬운 결과 해석이 가능한 CART가 사용하기 용이함을 보인다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.287-289
• /
• 2005

This paper presents a hybrid neural network system which used a Self-Organizing Map and Multilayer Perceptron for the problem of Malay syllables speech recognition. The novel idea in this system is the usage of a two-dimension Self-organizing feature map as a sequential mapping function which transform the phonetic similarities or acoustic vector sequences of the speech frame into trajectories in a square matrix where elements take on binary values. This property simplifies the classification task. An MLP is then used to classify the trajectories that each syllable in the vocabulary corresponds to. The system performance was evaluated for recognition of 15 Malay common syllables. The overall performance of the recognizer showed to be 91.8%.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.5
• /
• pp.1-11
• /
• 1994

In computer graphics since objects atre constructed by lines and curves, the high-speed curve generator is indispensible for computer aided design and simulatation. Since the functions of graphic generation can be represented as a series of matrix operations, in this paper, two kind of the high-speed Bezier curve generator that uses matrix equation and a recursive relation for Bezier polynomials are designed. And B-spline curve generator is designed using interdependence of B-spline blending functions. As the result of the comparison of designed curve generator and reference [5], [6] in the operation time and number of operators, the curve generator with 1-dimensional systolic array processor for matrix vector operation that uses matrix equation for Bezier curve is more effective.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.4
• /
• pp.234-239
• /
• 2014

This paper presents an empirical study on selecting a small amount of useful unlabeled data to improve the classification accuracy of semi-supervised learning algorithms. In particular, a hybrid method of unifying the simply recycled selection method and the incrementally-reinforced selection method was considered and evaluated empirically. The experimental results, which were obtained from well-known benchmark data sets using semi-supervised support vector machines, demonstrated that the hybrid method works better than the traditional ones in terms of the classification accuracy.

• Journal of the Korea Society of Computer and Information
• /
• v.7 no.4
• /
• pp.121-126
• /
• 2002

The stereoscopic image is that we can see it closer than a real thing compared to 2D image, and it has influence on human's vision information because it is more natural method to feel connections between the spaces of the image and himself. There are several method convert from 2d image to 3d image. But, in this paper, we are propose the image separate algorithm of continuous input system through a spatial analysis, not be done with 2D still image. Additionally, we will adapt to the moving vector which has been used in MPEG. In this experiment, we obtained the effect of 3D image.

• Journal of information and communication convergence engineering
• /
• v.16 no.3
• /
• pp.148-152
• /
• 2018

Adversarial attacks on artificial intelligence (AI) systems use adversarial examples to achieve the attack objective. Adversarial examples consist of slightly changed test data, causing AI systems to make false decisions on these examples. When used as a tool for attacking AI systems, this can lead to disastrous results. In this paper, we propose an ensemble of degraded convolutional neural network (CNN) modules, which is more robust to adversarial attacks than conventional CNNs. Each module is trained on degraded images. During testing, images are degraded using various degradation methods, and a final decision is made utilizing a one-hot encoding vector that is obtained by summing up all the output vectors of the modules. Experimental results show that the proposed ensemble network is more resilient to adversarial attacks than conventional networks, while the accuracies for normal images are similar.

• International Journal of CAD/CAM
• /
• v.6 no.1
• /
• pp.117-123
• /
• 2006

Recently with the development of 3D modeling and digitizing tools, more and more models have been created, which leads to the necessity of the technique of 3D mode retrieval system. In this paper we investigate a new method for 3D model retrieval based on orthogonal projections. We assume that 3D models are composed of trigonal meshes. Algorithms process first by a normalization step in which the 3D models are transformed into the canonical coordinates. Then each model is orthogonally projected onto six surfaces of the projected cube which contains it. A following step is feature extraction of the projected images which is done by Moment Invariants and Polar Radius Fourier Transform. The feature vector of each 3D model is composed of the features extracted from projected images with different weights. Our System validates that this means can distinguish 3D models effectively. Experiments show that our method performs quit well.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2011.01a
• /
• pp.185-188
• /
• 2011

In order to estimate spatial charging process in the corona charging which has been used to make polymer electret, the electrical properties of polypropylene film were obtained from Thermally Stimulated Current(TSC) measurements after corona charging between knife electrode and cylinder electrode with the voltages of -5, -6, -7 and -8[kV], respectively. And then the electrostatic contour and the electric field vector were also simulated by using Finite Element Method(FEM). The edge effect around edge of knife electrode affected the electrostatic contour on surface of specimen and the electric field concentration inside specimen. The uneven charging state in the electret due to the mistake on design could be calculated and so the optimal design of corona charging device which is appropriate to various materials is come to be practicable.