• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04b
• /
• pp.891-894
• /
• 2001

본 논문은 서로 다른 얼굴이미지 사이의 얼굴매칭기법의 새로운 방법인 TBCC(T-Block constraints Condition)얼굴매칭기법을 제안한다. 본 논문에서 제안하는 방법은 T영역안의 이목구비에다가 제어영역을 두고, T영역과 제어영역들을 분리하여 2개의 영상으로 각각 만든 다음에 각각의 correspondence가 있는 영상끼리 와핑(Warping)을 한 후에, 제어영역들은 2단계구조를 가진 계층적인 선형조합(Linear Combination)모델에 적용시켜 최적의 위치를 찾아낸 후에, T영역에 와핑시켜서 하나의 합성사진을 만들어 낸다. 합성사진에서 피부색이 다른 문제는 정규분표를 이용한 크로스디졸브(Cross-Dissolve)방법인 이미지프로세싱 기법을 새롭게 적용하며, 그리고, T모양의 자국이 남는 것은 본 논문에서 제안하는 T-Block Color Interpolation방법을 적용해서 해결한다.

• Smart Media Journal
• /
• v.3 no.1
• /
• pp.16-21
• /
• 2014

The demand for food rises proportionally as population grows. To be able to achieve sustainable supply of livestock products, efficient farm management is a necessity. With the advancement in technology it also brought innovations that could be harness in order to achieve better productivity in animal production and agriculture. Precision Livestock Farming (PLF) is a budding concept of making use of smart sensors or available devices to automatically and continuously monitor and manage livestock production. With this concept, this paper introduces a swine management system that integrates image processing technique for weight monitoring. This system captures pig images using camera, evaluate and estimate the weight base on the captured image. It is comprised of Pig Module, Breeding Module, Health and Medication Module, Weighr Module, Data Analysis Module and Report Module to help swine farm administrators better understand the performance and situation of the swine farm. This paper aims to improve the management in both small and big livestock raisers.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.796-799
• /
• 2006

This paper presents a new method for automatic detection of skew in a document image using mathematical morphology. To speed up processing, we use reduced image but it still requires long time to estimate the skew angle so the proposed method works with region of interest, not with whole image. Character strings are connected by using morphological closing operation and a component labeling is used to select region of interest. The method considers the lowermost pixels of characters in candidate regions in the binary image of original document image. Experimental results shows that the proposed method is extremely fast and robust as well as independent of script forms.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.27 no.2
• /
• pp.173-182
• /
• 2007

Cracks in concrete structures should be measured to periodically assess potential problems in durability and serviceability. Conventional crack measurement systems depend on visual inspections and manual measurements of the crack features such as width, length, and direction using microscope and crack gage. However, conventional methods take long time as well as manpower, and lack quantitative objectivity resulted by inspectors. In this study, an evaluation technique for concrete surface cracks is developed using image processing and artificial neural network. Developed technique consists of three major parts: (1) crack detection (2) crack analysis and (3) pattern recognition. To examine validity of the technique developed in this study, crack analyzing tests were performed on the images obtained from various types of concrete surface cracks. The test results revealed that the system is highly effective in automatically analyzing concrete surface cracks in terms of features and patterns of cracks.

• KIEAE Journal
• /
• v.12 no.5
• /
• pp.77-84
• /
• 2012

Daylighting is the controlled admission of natural light into a space, reducing electric lighting and saving energy. By providing a direct link to the dynamic and perpetually evolving patterns of outdoor illumination, daylighting helps create a visually stimulating and comfort environment for building occupants, while reducing energy costs. Especially, however, glare is the most important factor in daylighting, which is issued by incoming direct sunlight into windows. This study analyzed the discomfort glare on a daylighting window by using Image processing methodology and found a solution to problems with glare source of occupants. There are several ways to evaluate discomfort glare such as UGR (Unified Glare Rating) of CIE, DGI (Daylight Glare Index, Hopkinson, 1972) and VCP (Visual Comfort Probability) of IES. These are used to apply to the relatively little artificial light source and they cannot cover discomfort glare from a real daylighting window. In this regarding, this paper aimed to calculate DGI index of the real daylighting window in a experimental space by using image processing methodology. The variables and outcomes are luminance distribution of non-shading window, effect of venetian blind installed on the window and locations related to position index of DGI.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.429-432
• /
• 2020

본 논문에서는 카메라의 포커싱과 아웃포커싱에 의해 이미지에서 뿌옇게 표현되는 DoF(Depth of field, 피사계 심도) 영역을 합성곱 신경망을 통해 찾는 방법을 제안한다. 우리의 접근 방식은 RGB채널기반의 상호-상관 필터를 이용하여 DoF영역을 이미지로부터 효율적으로 분류하고, 합성곱 신경망 네트워크에 학습하기 위한 데이터를 구축하며, 이렇게 얻어진 데이터를 이용하여 이미지-DoF가중치 맵 데이터 쌍을 설정한다. 학습할 때 사용되는 데이터는 이미지와 상호-상관 필터 기반으로 추출된 DoF 가중치 맵을 이용하며, 네트워크 학습 단계에서 수렴률을 높이기 위해 스무딩을 과정을 한번 더 적용한 결과를 사용한다. 본 논문에서 제안하는 합성곱 신경망은 이미지로부터 포커싱과 아웃포커싱된 DoF영역을 자동으로 추출하는 과정을 학습시키기 위해 사용된다. 테스트 결과로 얻은 DoF 가중치 이미지는 입력 이미지에서 DoF영역을 빠른 시간 내에 찾아내며, 제안하는 방법은 DoF영역을 사용자의 ROI(Region of interest)로 활용하여 NPR렌더링, 객체 검출 등 다양한 곳에 활용이 가능하다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.22 no.6
• /
• pp.587-596
• /
• 2009

The fiber bridging model is the crucial factor to predict or analyze the tensile behavior of fiber reinforced cementitious composites. This paper presents the fiber bridging constitutive law considering the distribution of fiber inclined angle and the number of fibers in engineered cementitious composites. The distribution of fiber inclined angle and the number of fibers are measured and analyzed by the image processing technique. The fiber distribution are considerably different from those obtained by assuming two- or three-dimensional random distributions for the fiber inclined angle. The simulation of the uniaxial tension behavior was performed considering the distribution of fiber inclined angle and number of fibers measured by the sectional image analysis. The simulation results exhibit multiple cracking and strain hardening behavior that correspond well with test results.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1993.04b
• /
• pp.92-96
• /
• 1993

Modern NC machine tools require a more and more exact monitoring and control of cutting process. A very important parameter is tool wear used to optimize cutting data and to change tools at the right time. In this study, a real-time multi-tool fracture method is proposed for measuring crater from fringe patterns generated by uni-shadow-moire. An optical technique using laser for non-contact measurement of tool fracture is presented. The technology in which the tool is illuminated by a beam of He-Ne laser and then the image of tool fracture CCD camera.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2009.05a
• /
• pp.491-492
• /
• 2009

This paper presents evaluation results of fiber dispersion, number of fibers in a unit area, fiber orientation according to the mix-proportion of ECC and bridging stress calculated on the basis of fiber distribution characteristics.

• Korean Journal of Construction Engineering and Management
• /
• v.18 no.2
• /
• pp.21-29
• /
• 2017

Recently, with having a great interest of the general public for the AR (Augmented Reality) technology, there have been lots of study to improve efficiency of a construction equipment with applying the AR technology to a construction equipment. The clear extrinsic calibration is essential to applying AR technology at the construction site without any error which came from superimposition between 'Real world' and 'Virtual world'. However, on the construction site, the clear extrinsic calibration is not possible, because of lack of time and budget for the specific survey, also, the huge error of the outdoor tracking system such as gyro, GPS system and so on. In this study, we do research about seamless superposition with unclear extrinsic calibration and the image process method for making AR navigator operating in the excavator. Based on this study, we figure that we can fully develop the AR navigator for the excavator. Furthermore, thereby operating AR navigator at many construction sites, we expect that the efficiency of the excavator will be improved. In addition, we can develop AR navigator for not only a excavator but all about construction equipment.