• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.169-171
• /
• 2003

현실감 있는 얼굴 모델을 생성하기 위한 방법은 70년대부터 계속되었지만, 얼굴 구조의 복잡성, 색상과 텍스처의 다양한 분포, 잔주름과 같은 미세한 부분을 표현하기 어렵다는 정들로 인해 아직까지도 계속 연구되어지고 있다. 본 논문은 기존의 하드웨어 의존적인 3차원 얼굴 모델을 생성 방법이 아닌 2차원 얼굴 영상만으로 얼굴 모델을 생성하는 방법을 제시한다. 연구 수행 단계는 크게 얼굴 영역 검출 과정과 얼굴 모델링 과정으로 나뉘어지며, 얼굴 영역 검출을 위해 정규화된 TS 색상값과 얼굴의 피부색에 대한 평균과 공분산을 이용한 마할라노비스 거리 측정법을 이용한다. 얼굴 모델링 과정에서는 2차원 영상으로부터 3차원 정보를 추출한 뒤 일반 얼굴 모델에 변형을 주어 모델을 생성한다. 보다 현실감 있는 모델을 생성하기 위해 텍스쳐 매핑 기법을 추가한다. 본 연구를 통해 생성되는 얼굴 모델은 아바타 생성, 화상회의, 인증 시스템과 같은 분야에 적용 가능하며, 입력 영상에 대한 제약점을 줄이고 또한 사람의 손이 거치지 않고 전체적으로 자동화되어 처리할 수 있는 시스템을 제안한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.185-188
• /
• 2015

미리 저장된 객체의 3차원 특징점(Feature point) 좌표와 카메라 영상의 2차원 특징점 좌표를 매칭(Matching)하여 객체를 추적하는 방식의 경우, 카메라의 시점이 변할 때 특징점에서 발생되는 원근 효과(Perspective effect)가 반영되지 못하여 특징점 매칭 오류가 발생한다. 따라서 특징점에서 발생하는 원근 효과를 반영하여 정확한 카메라 포즈를 추정하기 위해 이전 프레임(Frame)의 카메라 포즈(Camera Pose)에 맞추어 텍스쳐가 포함 된 3차원 객체의 모델을 렌더링 하여 원근 효과를 적용한 후, 현재 카메라 영상과 특징점 매칭하여 프레임 사이의 카메라 움직임을 구하여 객체를 추적한다. 더 나아가 본 논문에서는 특징점 매칭에서 발생하는 작은 오류들로 인한 미세한 카메라 움직임은 2단계의 임계치(Threshold)를 적용하여 떨림 현상으로 간주하여 떨림 현상이 제거된 객체 추적을 수행한다. 매 프레임마다 카메라 포즈에 맞춘 추적 객체를 렌더링 하기 때문에 떨림 현상으로 간주되어 제거된 카메라 움직임은 누적되지 않고, 추적 오류도 발생시키지 않는다.

• Korean Journal of Food Science and Technology
• /
• v.21 no.3
• /
• pp.338-344
• /
• 1989

This study was conducted to investigate the mechanism involved with gelation of soybean proteins, 7S and 11S. For the preparations of soybean gels, calsium coagulation and isoelectric point precipitation through the lactic acid fermentation were employed. The textural properties and microstructure of soybean curds were examined by Instron Universal Testing Machine and Scanning Electron Microscope(SEM), respectively. Soybean cheeses were also prepared from soyprotein curds. The characteristics of prepared soybean cheeses were studied by Instron and Sensory evaluation. Microstructure of soybean curds demonstrated by SEM differed markely, postulating that molecular interaction occured in the curds varied with type of protein and coagulative conditions. Textural parameter measured by Instron demonstrated that the curds and the cheeses made through lactic acid fermentation showed higher values in hardness, gumminess and chewiness than those coagulated with $CaCl_2$ 11S PRF could give the curds with higher values in hardness, cohesiveness, springiness, gumminess and chewiness than SPI and 7S PRF Sensory evaluation results showed that soybean cheese made from 11S PRF scored higher values in taste, chewiness, and hardness. However, panels preferred soybean cheese prepared from SPI in color, chewiness and brittleness.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.393-399
• /
• 2021

In this study, experiments were conducted to estimate the PM concentrations by learning the nighttime CCTV images of various PM concentrations environments. In the case of daytime images, there have been many related studies, and the various texture and brightness information of images is well expressed, so the information affecting learning is clear. However, nighttime images contain less information than daytime images, and studies using only nighttime images are rare. Therefore, we conducted an experiment combining nighttime images with non-uniform characteristics due to light sources such as vehicles and streetlights and building roofs, building walls, and streetlights with relatively constant light sources as an ROI (Region of Interest). After that, the correlation was analyzed compared to the daytime experiment to see if deep learning-based PM concentrations estimation was possible with nighttime images. As a result of the experiment, the result of roof ROI learning was the highest, and the combined learning model with the entire image showed more improved results. Overall, R² exceeded 0.9, indicating that PM estimation is possible from nighttime CCTV images, and it was calculated that additional combined learning of weather data did not significantly affect the experimental results.

• Journal of Korea Multimedia Society
• /
• v.12 no.5
• /
• pp.737-744
• /
• 2009

This paper is to study and implement 3D Virtual Face Aesthetic Surgery System which provides more satisfaction by comparing the before-and-after plastic face surgery using 3D face model. For this study, we implemented 3D Face Model Generating System which resembles 2D image of the user based on 3D Korean standard face model and user's 2D pictures. The proposed 3D Virtual Face Aesthetic Surgery System in this paper consists of 3D Face Model Generating System, 3D Skin Texture Mapping System, and Detailed Adjustment System for reflecting the detailed description of face. The proposed system provides more satisfaction to the medical uses and stability in the surgery in compare with other existing systems.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.02a
• /
• pp.376-377
• /
• 2011

결정질 실리콘 태양전지 연구에 있어서 가장 중요한 부분은 재료의 저가화와 공정의 단순화에 의한 저가의 태양전지 셀 제작 부분과 고효율의 태양전지 셀 제작 부분이다. 본 논문에서는 마이크로 블라스터를 이용하여 폐 실리콘 웨이퍼를 태양전지용 재생웨이퍼를 제작함으로써 고효율을 가지는 단결정 실리콘 웨이퍼를 저 가격에 생산하기 위한 것이다. 특히 마이크로 블라스터를 이용하여 폐 실리콘 웨이퍼를 가공 할 때 표면에 생성되는 요철은 기존 태양전지 셀 제작에서 텍스쳐링 공정과 같은 표면 구조를 가지게 됨으로써 태양전지 셀에 제작 공정을 줄일 수 있는 효과도 가지게 된다. 마이크로 블라스터는 챔버 내에 압축된 공기나 가스에 의해 가속 된 미세 파우더들이 재료와 충돌하면서 재료에 충격을 주고 그 충격에 의해 물질이 식각되는 기계적 건식 식각 공정 기술이다. 이러한 물리적 충격을 이용하는 마이크로 블라스터 공정은 기존 재생웨이퍼 제작 공정 보다 낮은 재처리 비용으로 간단하게 태양전지용 재생웨이퍼를 제작 할 수 있다. 하지만 마이크로 블라스터를 이용하면 표면에 식각된 미세 파티클의 재흡착이 일어나게 되므로 이를 제거하기 위하여 DRE(damage remove etching) 공정이 필요하게 된다. 본 연구에서는 이방성, 등방성 식각 공정으로 태양전지용 재생웨이퍼를 제작하기 위해 가장 적합한 DRE 공정을 찾기 위해 등방성 식각은 RIE 식각으로, 그리고 이방성 식각은 TMAH 식각을 이용하였다. 마이크로 블라스터 공정 후 표면 반사율과 SEM 사진을 이용한 표면 요철 구조를 확인 하였고, DRE 공정 후 표면 반사율과 SEM 사진을 이용하여 표면 요철 구조를 확인 하였다. 각각의 lifetime을 측정하여 표면 식각으로 생성된 결함들을 분석하여 태양전지용 재생웨이퍼 제작에 가장 적합한 공정을 확인 하였다.

• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.363-367
• /
• 2009

Recently, the manufacturing of micro-cavity by means of laser surface texturing (LST) technique and low friction study by the LST have been in great progress. Most of current works have been dealing with the effect of cavity on friction and wear. The main objective of the present study was to investigate numerically two-dimensional lubrication characteristics of micro-dimple shapes fabricated on solid surfaces, and this study utilized the commercial CFD code (Fluent V.6.3). For the evaluation, preliminary simulation was conducted and numerical predictions were compared with the analytic solution obtained from the Reynolds's equation. Mainly, the present study investigated the influence of dimple depth, pattern shapes, and film thickness on lubrication characteristics related to the reduction of friction. It is found that the existence of micro-dimpled surface makes it possible to substantially reduce the friction forces exerted on the surfaces. In particular, substantial decrease in shear stresses was observed as the lubricant film thickness decreases. For instance, in the case of the film thickness of 0.01 mm, the estimated shear stress decreases up to about 40%. It indicates that the film thickness would be important factor in designing the micro-dimpled surfaces. Furthermore, it was observed that such a optimum dimple depth would be present because the dimple depth larger than the optimum value did no longer affect the reduction in shear stresses.

• Journal of the Korean Society for Precision Engineering
• /
• v.31 no.5
• /
• pp.381-387
• /
• 2014

This paper introduces a system to machine micro-sized patterns effectively on surface based on micro-milling process using tools with simultaneous rotation and oscillation, oscillation micro milling. To review the effectiveness of proposed concept, we integrated a micro-spindle supported by active magnetic bearings with a precision 3-axis air bearing stage using double-wedge mechanism, and tested this oscillation milling. Two types of oscillation milling were tested, which are linear oscillation milling with a flat end mill and elliptical oscillation milling with a ball end mill with 0.3 mm of diameter. The spindle was rotating 110 krpm and workpiece was moving constant speed of 2~8 mm/sec during the oscillation milling. As the results, multiple oval shape dimples were generated in regular spacing, and the variation of elliptical motion made different shapes of patterns. The results showed that proposed oscillation milling can be successfully used for machining repeated micro-patterns.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.6
• /
• pp.89-101
• /
• 2014

In this paper, we proposed a new local micro pattern, Signed Local Directional Pattern(SLDP). SLDP uses information of edges to represent the face's texture. This can produce a more discriminating and efficient code than other state-of-the-art methods. Each micro pattern of SLDP is encoded by sign and its major directions in which maximum edge responses exist-which allows it to distinguish among similar edge patterns that have different intensity transitions. In this paper, we divide the face image into several regions, each of which is used to calculate the distributions of the SLDP codes. Each distribution represents features of the region and these features are concatenated into a feature vector. We carried out facial expression recognition with feature vectors and SVM(Support Vector Machine) on Cohn-Kanade and JAFFE databases. SLDP shows better classification accuracy than other existing methods.

• Korean Journal of Food Science and Technology
• /
• v.19 no.6
• /
• pp.480-485
• /
• 1987

Quick cooking rice was produced by cooking raw, white long grain rice with three different methods which were atmospheric cooking in electric cooker, autoclaving and precooking followed by autoclaving. The drying rate curve showed that cooking methods affected little the drying pattern of the cooked rice except that precooking treatment before autoclaving retarded dehydration. The quick cooking rice prepared by precooking and autoclaving method among three cooking methods showed the fastest rehydration rate and took 7.5 minutes in reaching equilibrium moisture content by soaking in boiling water. And it had less hard and more cohesive texture than the milled rice cooked by conventional method and the quick cooking rice prepared by electric cooking when reconstituted. The microstructure of quick cooking rice was investigated among cooking methods.