• 한국데이터정보과학회:학술대회논문집
• /
• 2002.06a
• /
• pp.35-50
• /
• 2002

본 연구에서는 ECHAM4/OPYC3 자료를 이용하여 관심영역인 동아시아 영역과 기후변화의 신호가 강할 것으로 추정되는 시베리아 영역에 대하여 베이즈 접근법을 적용하여 기후변화 신호를 탐지하고, 시나리오 자료에 대하여 기후변화 원인별 정량화를 실시하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.574-576
• /
• 1998

본 논문은 디지털 칼라 영상 정보를 보호하기 위해 위치와 주파수 정보를 가지는웨이브릿 변환을 이용한 워터마킹(watermarking)방법에 대하여 연구하였다. 본 연구에서는 칼라 영상의 각 채널에 대해 인간의 시각에 덜 민감한 고주파 계수영역과 영상 압축에 견고한 저주파 계수영역에 워터마크를 삽입하였다. 이 방법은 불법적인 조작에 강한 워터마크를 구현할 수가 있다. 실험에서 칼라 영상의 각 채널을 웨이브릿 변환한 후 Seal 영상을 삽입하였다. 실험 결과, 워터마크된 칼라 영상에 삽입된 Seal 영상은 잡음으로 영상이 훼손된 경우라도 Seal 영상을 AND 연산함으로써 구할 수 있었고, 압축에도 견고함을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.83-86
• /
• 2003

인터넷의 발달과 네트워크의 발달로 인하여 디지털 데이터에 대한 저작권 문제가 발생하고 있다. 최근 디지털 데이터에 대한 문제점을 해결하기 위한 방법으로 디지털 워터마킹에 대한 연구가 많이 이루어지고 있다. 본 논문에서는 디지털 데이터의 무단 복제와 배포를 박기 위해 웨이브렛을 이용한 주파수 변환을 이용하고, 워터마크로 디지털 지문을 삽입하였다. 웨이브렛 변환 영역중에 HL영역에 워터마크가 가시성과 추출 결과에 좋은 결과가 있었다.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1320-1322
• /
• 1996

기존의 창함수를 이용한 1차원 FIR 필터의 설계는 설계방법이 단순해 비교적 많이 사용되고 있으나 통과영역과 천이영역을 가변할 수 없다. 본 연구에서는 두 개의 영역을 가변할 수 있는 창함수를 사용하여 1차원 임펄스 응답을 구한 후 McCllenan 변환 방법을 이용하여 2차원 FIR 필터를 실현하여 특성을 비교하여 2차원 FIR 필터의 설계에 적용할 수 있음을 밝혔다.

• Journal of the Korean Society of Industry Convergence
• /
• v.3 no.4
• /
• pp.299-306
• /
• 2000

이 논문은 파이프라인 함성을 위한 상위수준 데이터 경로 하성과 콘트롤 합성의 통합에 관한 연구이다. 현재 대부분의 상위수준 합성 방법은 콘트롤 영역의 영향을 무시하는데 보다 나은 설계를 위하여 데이터 경로디자인 영역과 콘트롤 디자인 영역을 통합하여 탐색하는 파이프라인 상위수준함성 도구를 구현했다. 이 도구는 비용 제한 하에서 최고 성능의 파이프라인을 합성하는 비용재한합성과 성능 제한 하에서 최서 비용의 파이프라인을 합성하는 성능 제한합성의 두 가지 방식을 제공한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.50-52
• /
• 2011

영상 관제 시스템을 구성할 때 카메라 감시영역 간의 중첩된 영역이 발생한다. 이러한 중첩된 영역을 설정해놓고 카메라를 서로 연동시킬 수 있다. 본 연구에서는 와이드뷰 카메라와 PTZ 카메라의 연동에 의한 실시간 오브젝트 트래킹 시스템을 구현하였다. 본 연구의 결과는 향후 다수의 카메라의 연동으로 확장될 수 있다.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.4
• /
• pp.189-198
• /
• 2024

Lipreading is one of the important parts of speech recognition, and several studies have been conducted to improve the performance of lipreading in lipreading systems for speech recognition. Recent studies have used method to modify the model architecture of lipreading system to improve recognition performance. Unlike previous research that improve recognition performance by modifying model architecture, we aim to improve recognition performance without any change in model architecture. In order to improve the recognition performance without modifying the model architecture, we refer to the cues used in human lipreading and set other regions such as chin and cheeks as regions of interest along with the lip region, which is the existing region of interest of lipreading systems, and compare the recognition rate of each region of interest to propose the highest performing region of interest In addition, assuming that the difference in normalization results caused by the difference in interpolation method during the process of normalizing the size of the region of interest affects the recognition performance, we interpolate the same region of interest using nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation, and compare the recognition rate of each interpolation method to propose the best performing interpolation method. Each region of interest was detected by training an object detection neural network, and dynamic time warping templates were generated by normalizing each region of interest, extracting and combining features, and mapping the dimensionality reduction of the combined features into a low-dimensional space. The recognition rate was evaluated by comparing the distance between the generated dynamic time warping templates and the data mapped to the low-dimensional space. In the comparison of regions of interest, the result of the region of interest containing only the lip region showed an average recognition rate of 97.36%, which is 3.44% higher than the average recognition rate of 93.92% in the previous study, and in the comparison of interpolation methods, the bilinear interpolation method performed 97.36%, which is 14.65% higher than the nearest neighbor interpolation method and 5.55% higher than the bicubic interpolation method. The code used in this study can be found a https://github.com/haraisi2/Lipreading-Systems.

• Journal of Service Research and Studies
• /
• v.10 no.4
• /
• pp.89-100
• /
• 2020

The smart factory is an important axis of the 4th industrial revolution. Smart factory is a system that induces the maximum efficiency and effectiveness of production using the IoT and intelligent sensing systems. The product lifecycle management technique is a method that can actively reflect the consumer's requirements in the smart factory and manage the entire process from the consumer to the post management. There have been many studies on product lifecycle management, but studies on how to organize product lifecycle management knowledge domains in preparation for the era of the 4th industrial revolution were insufficient. This study analyzed the opinions of a group of experts preparing for the 4th industrial revolution in terms of product lifecycle management. The impact of the 4th industrial revolution on the detailed knowledge areas of product lifecycle management was investigated. The changes in product lifecycle management were summarized using a qualitative data analysis technique for a group of experts. Based on the opinions of experts, the product lifecycle management, which consists of a total of 30 detailed knowledge areas, was prepared to supplement or prepare for the 4th industrial revolution. This study investigates changes in product lifecycle management in preparation for the 4th industrial revolution in the knowledge domain of the existing defined product life cycle management. In future research, it is necessary to redefine the knowledge domain of product life cycle management suitable for the era of the 4th industrial revolution and investigate the perception of experts. Considering the social culture and technological change factors of the 4th industrial revolution, the scope and scope of product life cycle management can be newly defined.

• Journal of Internet Computing and Services
• /
• v.11 no.3
• /
• pp.43-51
• /
• 2010

In this paper, we evaluate the system throughput and the method of optimal wireless resource allocation for the access zone (AZ) and relay zone (RZ) in downlink when the cell coverage is extended using the non-transparent Relay Station (RS) in a 2-hop cellular relay system based on IEEE802.16j, which uses the OFDMA-TDD structure. For the analyses, we first introduce the MAC (Media Access Control) frame structure in the non-transparent mode, then we investigate the interfering elements in both AZ and RZ for the network devices such as the Mobile Station (MS) and RS. Through computer simulation, we analyze the cell coverage extension and system throughput in terms of the distance between Base Station (BS) and RS, then we present the amount of the optimal allocation of wireless resource for the AZ and RZ in downlink using our results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.2
• /
• pp.358-363
• /
• 2009

Text region detection from the natural scene images used in a variety of applications, many research are needed in this field. Recent research methods is to detect the text region using various algorithm which it is combination of edge based and connected component based. Therefore, this paper proposes an text region detection using edge and regional minima/maxima transformation algorithm from natural scene images, and then detect the connected components of edge and regional minima/maxima, labeling edge and regional minima/maxima connected components. Analysis the labeled regions and then detect a text candidate regions, each of detected text candidates combined and create a single text candidate image, Final text region validated by comparing the similarity and adjacency of individual characters, and then as the final text regions are detected. As the results of experiments, proposed algorithm improved the correctness of text regions detection using combined edge and regional minima/maxima connected components detection methods.