• The tire
• /
• s.193
• /
• pp.47-59
• /
• 1997

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.2
• /
• pp.31-43
• /
• 2000

This paper presents a novel algorithm lot extraction of the eye and mouth fields (facial features) from 2D gray level face images. First of all, it has been found that Eigenfeatures, derived from the eigenvalues and the eigenvectors of the binary edge data set constructed from the eye and mouth fields are very good features to locate these fields. The Eigenfeatures, extracted from the positive and negative training samples for the facial features, ate used to train a MultiLayer Perceptron(MLP) whose output indicates the degree to which a particular image window contains the eye or the mouth within itself. Second, to ensure robustness, the ensemble network consisting of multiple MLPs is used instead of a single MLP. The output of the ensemble network becomes the average of the multiple locations of the field each found by the constituent MLPs. Finally, in order to reduce the computation time, we extracted the coarse search region lot eyes and mouth by using prior information on face images. The advantages of the proposed approach includes that only a small number of frontal faces are sufficient to train the nets and furthermore, lends themselves to good generalization to non-frontal poses and even to other people's faces. It was also experimentally verified that the proposed algorithm is robust against slight variations of facial size and pose due to the generalization characteristics of neural networks.

• Cartoon and Animation Studies
• /
• s.13
• /
• pp.97-114
• /
• 2008

This study aims to study mouth shapes suitable to the shapes of Korean consonants and vowels for Korean animations by analyzing the process of English-speaking animation lip-sync based on pre-recording in the United States. A research was conducted to help character animators understand the concept of Korean lip-sync which is done after recording and to introduce minimum, basic mouth shapes required for Korean expressions which can be applied to various characters. In the introduction, this study mentioned the necessity of Korean lip-sync in local animations and introduced the research methods of Korean lip-sync data based on English lip-sync data by laking an American production as an example. In the main subject, this study demonstrated the characteristics and roles of 8 basic mouth shapes required for English pronunciation expressions, left out mouth shapes that are required for English expressions but not for Korean expressions, and in contrast, added mouth shapes required for Korean expressions but not for English expressions. Based on these results, this study made a diagram for the mouth shapes of Korean expressions by laking various examples and made a research on how mouth shapes vary when they are used as consonants, vowels and batchim. In audition, the case study proposed a method to transfer lines to the exposure sheet and a method to arrange mouth shapes according to lip-sync for practical animation production. However, lines from a Korean movie were inevitably used as an example because there has not been any precedents in Korea about animation production with systematic Korean lip-sync data.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.40 no.1
• /
• pp.34-42
• /
• 2012

This study aims to suggest the secure indicator of parks and green areas according to the properties of the location. There were 31 study areas where negotiation was finalized from 2001 to 2008. The location types were divided by land use type before development and the ratio of mountainous and constructed parks, the ratio of green change and so forth were analyzed according to the those types. As a result, location types were classified into mountainous type, arable land type, urban district type. In terms of locational property, even though the ratio of parks and green areas was the highest at 24.9% in arable land type, the ratio was not much different from the others. Mountainous type depends on the ratio of mountainous park with 10.5%, and arable land type, and urban district type has high ratio of constructed parks. And, the ratio of green change arable land type is degraded by 32.8% and the ratio of mountainous green change is serious in the mountainous type. As a result of the selection of negotiation indicator, arable land type is closely related to the ratio of parks and green areas and the ratio of green change; mountainous type, the ratio of park and green, the ratio of mountainous green change and the ratio of constructed park and green; urban district type, the ratio of green areas.

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.1
• /
• pp.26-32
• /
• 2009

The multi-level storage architecture has been widely adopted in servers and data centers. However, while prefetching has been shown as a crucial technique to exploit sequentiality in accesses common for such systems and hide the increasing relative cost of disk I/O, existing multi-level storage studies have focused mostly on cache replacement strategies. In this paper, we show that prefetching algorithms designed for single-level systems may have their limitations magnified when applied to multi-level systems. Overly conservative prefetching will not be able to effectively use the lower-level cache space, while overly aggressive prefetching will be compounded across levels and generate large amounts of wasted prefetch. We design and implement a hierarchy-aware lower-level prefetching strategy called PMS(Prefetching strategy for Multi-level Storage system) that applicable to any upper level prefetching algorithms. PMS does not require any application hints, a priori knowledge from the application or modification to the va interface. Instead, it monitors the upper-level access patterns as well as the lower-level cache status, and dynamically adjusts the aggressiveness of the lower-level prefetching activities. We evaluated the PMS through extensive simulation studies using a verified multi-level storage simulator, an accurate disk simulator, and access traces with different access patterns. Our results indicate that PMS dynamically controls aggressiveness of lower-level prefetching in reaction to multiple system and workload parameters, improving the overall system performance in all 32 test cases. Working with four well-known existing prefetching algorithms adopted in real systems, PMS obtains an improvement of up to 35% for the average request response time, with an average improvement of 16.56% over all cases.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.1393-1400
• /
• 2006

정서를 측정하고 나타내는 기술이 발전에 따라 문화적 보편성을 가진 얼굴표정 연구의 필요성이 증가하고 있다. 그리고 지금까지의 많은 얼굴 표정 연구들은 정적인 얼굴사진 위주로 이루어졌다. 그러나 실제 사람들은 단적인 얼굴표정만으로 정서를 인식하기 보다는 미묘한 표정의 변화나 얼굴근육의 움직임 등을 통해 정서상태를 추론한다. 본 연구는 동적인 얼굴표정이 정적인 얼굴표정 보다 정서상태 전달에서 더 큰 효과를 가짐을 밝히고, 동적인 얼굴 표정에서의 눈과 입의 정서인식 효과를 비교해 보고자 하였다. 이에 따라 15 개의 형용사 어휘에 맞는 얼굴 표정을 얼굴전체, 눈, 입의 세 수준으로 나누어 동영상과 스틸사진으로 제시하였다. 정서 판단의 정확성을 측정한 결과, 세 수준 모두에서 동영상의 정서인식 효과가 스틸사진 보다 유의미하게 높게 나타나 동적인 얼굴 표정이 더 많은 내적정보를 보여주는 것을 알 수 있었다. 또한 얼굴전체-눈-입 순서로 정서인식 효과의 차이가 유의미하게 나타났으며, 부정적 정서는 눈에서 더 잘 나타나고 긍정적 정서는 입에서 더 잘 나타났다. 따라서 눈과 입에 따른 정서인식이 정서의 긍정성-부정성 차원에 따라 달라짐을 볼 수 있었다.

• Journal of the Korean Ceramic Society
• /
• v.41 no.2
• /
• pp.131-135
• /
• 2004

The effect of Co additive on the microstructural evolusion of WC was investigated. A small amount of Co powder was placed on the top-center of the pure WC powder compact and then sintered at 1950$^{\circ}C$. During sintering some abnormally large WC grains of different size and shape observed depending on the distance from the liquid source. However, in the region far away from Co liquid source, it showed low densification and the grains of WC were very small and uniform in size. A small amount of Co liquid phase has a remarkable influence on the AGG of WC and it has been explained in terms of 2-D nucleation and growth mechanism.

• Proceedings of the KWS Conference
• /
• 2009.11a
• /
• pp.60-60
• /
• 2009

본 연구는 보론이 첨가된 저합금강 용접 열영향부에서의 보론 편석 거동 연구를 위해 보론이 10ppm 첨가된 저합금강을 이용하여 다양한 용접 입열량 및 외부 응력에 따른 용접부 CGHAZ의 보론 편석거동을 분석하였다. 이를 위해 Gleeble 시스템을 이용하여 다양한 입열량에 따른 CGHAZ를 열 및 열-응력 사이클을 통하여 재현하였다. 재현된 시편의 미세조직은 OM을 통하여 분석하고, 보론의 편석거동을 SIMS와 PTA 분석법을 통하여 분석하였다. 그 결과 입열량에 따른 보론의 편석 거동은 최초 입열량이 증가함에 따라 보론의 편석이 증가하다가 다시 감소하였는데 이는 비평형 편석 후 고온에서 유지시간이 길어짐에 따라 back diffusion 발생에 따른 영향으로 판단된다. 또한 외부 응력에 의한 보론 편석 거동 분석 결과, 용접 열 사이클 중 작용하는 외부 응력에 의해 결정립계 편석 감소하였는데 이는 외부 응력에 의한 오스테나이트 결정립 크기 감소에 따른 결정립계 증가의 영향으로 판단된다.

• Science of Emotion and Sensibility
• /
• v.1 no.1
• /
• pp.33-40
• /
• 1998

본 논문은 얼굴인식 분야에 있어서 필수 과정인 얼굴 및 얼굴의 주요소인 눈과 입의 추출에 관한 방법을 제시한다. 얼굴 영역 추출은 복잡한 배경하에서 움직임 정보나 색상정보를 사용하지 않고 통계적인 모델에 기반한 일종의 형찬정합 방법을 사용하였다. 통계적인 모델은 입력된 얼굴 영상들의 Hotelling변환 과정에서 생성되는 고유 얼굴로, 복잡한 얼굴 영상을 몇 개의 주성분 갑으로 나타낼 수 있게 한다. 얼굴의 크기, 영상의 명암, 얼굴의 위치에 무관하게 얼굴을 추출하기 위해서, 단계적인 크기를 가지는 탐색 윈도우를 이용하여 영상을 검색하고 영상 강화 기법을 적용한 후, 영상을 고유얼굴 공간으로 투영하고 복원하는 과정을 통해 얼굴을 추출한다. 얼굴 요소의 추출은 각 요소별 특성을 고려한 엣지 추출과 이진화에 따른 프로젝션 히스토그램 분석에 의하여 눈과 입의 경계영역을 추출한다. 얼굴 영상에 관련된 윤곽선 추출에 관한 기존의 연구에서 주로 기하학적인 모양을 갖는 눈과 입의 경우에는 주로 가변 템플릿(Deformable Template)방법을 사용하여 특징을 추출하고, 비교적 다양한 모양을 갖는 눈썹, 얼굴 윤곽선 추출에는 스네이크(Snakes: Active Contour Model)를 이용하는 연구들이 이루어지고 있는데, 본 논문에서는 이러한 기존의 연구와는 달리 스네이크를 이용하여 적절한 파라미터의 선택과 에너지함수를 정의하여 눈과 입의 윤곽선 추출을 실험하였다. 복잡한 배경하에서 얼굴 영역의 추출, 추출된 얼굴 영역에서 눈과 입의 영역 추출 및 윤곽선 추출이 비교적 좋은 결과를 보이고 있다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.5
• /
• pp.665-671
• /
• 2011

The Competitiveness of Hybrid hard disk drive(H-HDD) for solid state disk(SSD) comes from both lower power consumption and higher reading speed. This paper suggests a prefetching scheme that can improve the performance of Non-Volatile cache(NVCache) memory installed on the H-HDD through prefetching disk blocks as well as files to the NVCache. The proposed scheme makes the highly used data such as booting files copy to the NVCache as an unit of file and the frequently accessed blocks copy to the NVCache. This prefetching is done on the idle time of disk queue and the priorities of prefetched target blocks are based on both time and spatial locality of blocks. Experiments results show that the suggested method can improve response time of H-HDD and also lower the power consumption.