• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅳ
• /
• pp.2028-2031
• /
• 2003

This paper proposes a face identification algorithm, robust on lighting condition and complex background. The proposed method estimates facial area under bad light condition by expanding face color boundaries and then finds a lip using the templates for lips. Then the eyes are found using their topological relationship with the long and short axes of lip area. The experimental results have shown that the proposed algorithm is robust on lighting conditions and complex background.

• 한국독성학회:학술대회논문집
• /
• 한국독성학회 2005년도 춘계 국제심포지엄 및 학술대회
• /
• pp.111-111
• /
• 2005

• 복식문화연구
• /
• 제16권5호
• /
• pp.863-877
• /
• 2008

The purpose of this study is to investigate the effect of dot pattern size(0.8, 1.8, 2.5, 5, 8), color combination (BG/R, Y/B), value tone(lt/dk, p/g), area-ratio on image information. Sets of stimulus and response scales(7 point semantic) were used as experimental materials. The stimuli were 20 color pictures manipulated with the combination of dot pattern size, color combination, value tone and area-ratio using computer simulation. The subjects were 240 female undergraduates living in Gyeongsangnam-do. Image factor of the stimulus was composed of 4 different components, visibility, chastity.feminity, cuteness and attractiveness. In the visibility, color combination, value tone, area-ratio, dot pattern size showed independent effect. In the chastity feminity, color combination, value tone, showed independent effect. In the cuteness, value tone, area-ratio, dot pattern size showed independent effect. Significant interaction effects of color and area-ratio combination on visibility and cuteness were found. Interaction efforts of color and value tone combination, value tone and area-ratio was significant on cuteness. For visibility image, BG/R combination of color and yellow background/blue dots were effective. For cuteness image, pale/grayish tone and background/dots area-ratio were effective.

• 복식
• /
• 제58권6호
• /
• pp.54-68
• /
• 2008

The purpose of this study is to investigate the effect of dot pattern size(0.8, 1.8, 2.5, 5, 8), color combination(BG/R, Y/B), area-ratio on image formation. Sets of stimulus and response scales(7 point semantic) were used as experimental materials. The stimuli were 20 color pictures manipulated with the combination of dot pattern size, color combination, and area-ratio using computer simulation. The subjects were 240 female undergraduates living in Gyeongnam-do. Image factor of the stimulus was composed of 5 different components, visibility, attractiveness, cuteness, stability and high class image. In the cuteness, color combination, dot pattern size showed independent effect. In the stability, area-ratio, dot pattern size showed independent effect. Interaction effects of color and area-ratio combination was significant on cuteness. For visibility image 8cm yellow dot/blue background, for attractiveness image BG/R coloration, for cuteness image Y/B coloration and for stability image 0.8cm yellow dot/blue background were effective. According to the variation of dot pattern size, color combination and area-ratio, it was investigated that the images for a dress wearer were expressed diversely, were shown differently in image dimensions, and could be produced to different images.

• 전자공학회논문지
• /
• 제52권11호
• /
• pp.105-114
• /
• 2015

본 논문은 흉부 엑스레이 영상에서 배경 제거 및 관심 영역을 분할하는 기법을 제안한다. 일반적으로 화질 개선 기법을 적용할 때 영상의 밝기 정보나 주파수 정보를 이용하여 영상 선명도와 대비를 개선하는 방법을 사용한다. 이러한 기법을 엑스레이 영상 전체에 적용하는 경우 배경과 같은 영상의 불필요한 정보 때문에 좋은 성능을 얻기 어렵다. 그래서 본 논문은 사용자가 원하는 영역에만 화질 개선 기법을 적용할 수 있도록 배경 제거 및 관심 영역 (ROI)을 분할하는 방법을 제안한다. 배경 제거를 위해 먼저 원본 영상의 히스토그램 분포를 분석하고 문턱치 처리로 몸체와 배경을 일차적으로 분리한다. 다음으로 유도 필터 (guided filter)를 이용하여 몸체 경계 혹은 배경 경계를 보정한다. 관심 영역 분할을 위해서는 먼저 폐의 위치 정보를 이용하여 폐의 주 밝기 값을 찾는다. 이를 이용하여 문턱치 처리를 한 후 번호 매김과 상기 배경 정보를 이용하여 분류 이외의 것을 제거한다. 마지막으로 폐만 검출된 이진영상을 통해 경계 상자 영역을 생성한다. 모의실험을 통해 제안하는 기법의 우수성을 검증하였다.

• International Journal of Contents
• /
• 제9권1호
• /
• pp.6-10
• /
• 2013

Foreground estimation in object segmentation has been an important issue for last few decades. In this paper we propose a GrabCut based automatic foreground estimation method using block clustering. GrabCut is one of popular algorithms for image segmentation in 2D image. However GrabCut is semi-automatic algorithm. So it requires the user input a rough boundary for foreground and background. Typically, the user draws a rectangle around the object of interest manually. The goal of proposed method is to generate an initial rectangle automatically. In order to create initial rectangle, we use Gabor filter and Saliency map and then we use 4 features (amount of area, variance, amount of class with boundary area, amount of class with saliency map) to categorize foreground and background. From the experimental results, our proposed algorithm can achieve satisfactory accuracy in object segmentation without any prior information by the user.

• 한국조경학회지
• /
• 제19권4호
• /
• pp.58-79
• /
• 1992

This is a study of Sadaebu(previllaged class in a Yi-Dynasty士 大夫) clan villages in the Andong Area with emphasis on Formantion Process of development. The purposes and the contents of this study are as following : 1. To review the process and the background fo growth of Myung-Mun-Se-Ga(名文大家:the traditional famed & mighty clans) in Yi-Dynasty, in cases of Andong(安東) province. 2. To analyze about formation process of a sphere of their homogeneous clan villages, and to understand the factors affecting their sphere. 3. To identity the background factor of formation of their sphere of traditional clan villages in Andong province.

• 한국지능시스템학회논문지
• /
• 제18권6호
• /
• pp.797-807
• /
• 2008

본 논문은 복잡한 도로 영상에서 차량 검출의 효율성을 높이기 위해 체인코드를 이용한 차선의 검출로부터 도로 영역을 찾아 차량이 존재하는 차량 영역의 추출 기법을 제안한다 먼저, 복잡한 도로 영상에서 정확한 차선을 검출하기 위해 체인코드를 이용하여 에지 화소들간의 연결성을 고려한다. 주행 차량의 방향과 일치하는 차선을 검출한 후, 중앙의 차선으로부터 차도의 폭과 차선의 소실점을 찾아 인접하는 차도를 찾는다. 마지막으로 주행 차선과 인접 차선을 포함하는 도로 영역 내에 차량의 에지 정보를 이용하여 차량이 존재하는 차량 영역을 추출한다 따라서, 제안하는 차량 영역의 추출 기법은 복잡한 배경을 갖는 도로 영상에서 차량의 검출율을 높이고 추출된 차량 영역에 한정할 수 있기 때문에 차량을 검출하는데 매우 효율적이다. 본 논문은 제안하는 차량 영역의 추출 기법의 우수성을 복잡한 도로 영상에서 차량 검출율의 실험을 통해 검증하였다.

• 상하수도학회지
• /
• 제24권5호
• /
• pp.495-508
• /
• 2010

The aim of this study is to develop a quantified water loss Index to evaluate and manage leakage scientifically for the reduction of non-revenue water in water distribution systems. For the purpose, unavoidable background leakage suggested from UK water industry and IWA, and allowable water leakage in accord with the concept of allowable water loss are proposed by analyzing the inflow into two study water districts and the short-term water use of each customer in the districts. The study distribution areas are selected among the metered districts with good maintenance of leakage after improvement activities in Nonsan, medium sized city in Korea. Estimation models of allowable leakage are developed by metering and analyzing the minimum night flow at residential and commercial areas in the city. In the results of the investigation, it is estimated that background night flow in residential area was larger than that of commercial area where the types of business shows small water use characteristics. Meanwhile, night flow and background water loss on internal plumbing systems show great differences for each district which is influenced much by the water use characteristics and facilities scale. Based on metering water use data in various districts, leakage management criteria can be established under the consideration of domestic conditions in Korea by analyzing separated real water use and background leakage and it is possible to apply into presentation of optimal leakage level and reasonable time for working activities for leakage reduction.

• 제어로봇시스템학회논문지
• /
• 제16권3호
• /
• pp.269-276
• /
• 2010

In this paper, we present the detection method of moving objects based on two background models. These background models support to understand multi layered environment belonged in images taken by shaking camera and each model is MBM(Multiple Background Model) and TMBM (Temporal Median Background Model). Because two background models are Pixel-based model, it must have noise by camera movement. Therefore correlation coefficient calculates the similarity between consecutive images and measures camera motion vector which indicates camera movement. For the calculation of correlation coefficient, we choose the selected region and searching area in the current and previous image respectively then we have a displacement vector by the correlation process. Every selected region must have its own displacement vector therefore the global maximum of a histogram of displacement vectors is the camera motion vector between consecutive images. The MBM classifies the intensity distribution of each pixel continuously related by camera motion vector to the multi clusters. However, MBM has weak sensitivity for temporal intensity variation thus we use TMBM to support the weakness of system. In the video-based experiment, we verify the presented algorithm needs around 49(ms) to generate two background models and detect moving objects.