• 한국정보과학회논문지:소프트웨어및응용
• /
• 제35권10호
• /
• pp.600-607
• /
• 2008

이 논문은 방향성 2차원 타원형 필터(Multiple Oriented 2D Elliptical Filters;MO2DEFs)를 사용하여 스테레오 영상으로부터 포즈에 강인한 사람 검출을 제안한다. 기존의 물체 지향 크기 적응 필터(Object Oriented Scale Adaptive Filter;OOSAF)는 정면을 보고 있는 사람만을 검출하는 단점을 지니고 있는데 반해 제안한 방향성 2차원 타원형 필터는 사람의 크기나 포즈에 관계없이 사람을 검출하고 추적한다. 2D 공간-깊이 히스토그램에 특정 각도로 향하는 4개의 2차원 타원형 필터들을 적용하고, 필터링 된 히스토그램에서 임계값을 통해서 사람을 검출한 다음, MO2D2EFs 중 승적 결과가 가장 큰 2차원 타원형 필터의 방향을 사람의 방향으로 판단한다. 사람 후보들은 얼굴을 검출하거나 검출된 사람의 선택된 방향의 머리-어께 형태를 정합함으로서 검증한다. 실험 결과는 (1) 포즈 각도 예측의 정확도는 약 88%이고, (2) 제안한 MO2DEFs를 사용한 사람 검출의 성능이 OOSAF를 사용한 사람 검출의 성능보다 $15{\sim}20%$만큼 향상되었으며, 특히 정면이 아닌 사람의 경우에 더 향상이 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제7권4호
• /
• pp.46-51
• /
• 2002

본 논문은 얼굴 요소 중 눈썹을 검출하기 위한 기법으로, 눈썹은 얼굴 인식이나 표정 인식, 얼굴 애니메이션에 중요한 역할을 하는 요소이다. 색상 영역 분할을 통해 얼굴 영역을 검출한 다음, 형판 정합(template matching)을 통해 눈을 검출한다. 눈썹은 눈 바로 위에 위치하므로 검출된 눈의 위치 값을 이용하여 눈썹 후보 영역을 설정한다. 이렇게 설정된 눈썹 후보 영역에서 휘도(luminance) 성분의 히스토그램을 구한 다음, 이 히스토그램을 이용하여 thresholding 기법으로 눈썹을 검출한다. 일반적으로 이런 히스토그램은 하나의 bin을 갖는 peak나 valley가 무수히 많아 threshold 간을 결정하는데 어려움이 있다. 이런 어려움을 극복하고 좀더 쉽게 threshold 값을 찾기 위해 이런 peak나 valley를 제거해 히스토그램을 변형한다. 제안된 알고리즘은 얼굴 영역 검출부, 얼굴 요소 검출부, 그리고 FCP 추출부 등의 세 부분으로 구성되어있다.

• 한국의류산업학회지
• /
• 제9권2호
• /
• pp.145-150
• /
• 2007

Recently, active movements are occurring to reestablish our own traditional Korean images in all levels of society. One of them is about Ordinary Hanbok. Academics and business sectors are endeavoring to develop comfortable and Korean traditional Hanbok designs and popularize them as well. Based on the facts, the study focuses on developing Ordinary Hanbok designs for women by applying Korean traditional patchwork in order to present handmade, humane senses. First of all, the backgrounds, history and preference of Ordinary Hanbok are discussed through literatures and precedent studies as study methodology. Then, the study is analyzing the formative beauty of Korean traditional patchwork that contains emotions of women in the Chosun Dynasty. After that, it shows Ordinary Hanbok designs developed for women by using checks, free style, and yeoiju patterns of the fabrics dyed with clay and Chinese ink that recently attract people in particular. The results are like followings; First, the limits that excessive transformation of Hanbok may cause the lack of traditional beauty can be covered through using a variety of images of Korean traditional patchwork. Second, diverse face configuration of patchworks has very unique geometric forms so to be easily applied to modern Ordinary Hanbok designs. Third, the sewing technique of patchworks also plays a role as a design factor as important as face configuration. Fourth, utilizing fabric pieces may result in various unplanned face configuration and help resource recycling. Fifth, it is possible to coordinate Hanbok with other ordinary clothes that people normally wear on. Sixth, using patchworks images can encourage developing more active and refined designs added with traditional beauty. Seventh, natural dying and fabrics are eco-friendly, which is good for human health and environment. Finally, more studies should be done on design development based on precedent studies about Ordinary Hanbok. Also, it is important to foster social atmosphere and circumstances where people accept wearing Ordinary Hanbok as our natural culture.

• 한국항행학회논문지
• /
• 제15권6호
• /
• pp.1132-1137
• /
• 2011

본 논문에서는 사람의 신체 일부분을 추적하는 시스템을 위해서 피부영역을 추출하고 여러 개의 영역을 추적하는 다중 CAMShift 알고리즘(Multi Continuously Adaptive Mean Shift Algorithm)을 제안하였다. 입력 영상에서 피부영역을 추출하기 위해 영상의 RGB의 특정값을 기준으로 피부색에 적응적인 임계값을 적용하였다. 이때 적용된 피부영역을 양손, 얼굴 등에 초기 윈도우를 설정하였다. 이 영역들을 추적함에 있어 영역들 사이에 폐색 영역을 회피하기 위해 가우시안 배경 모델(Gaussian Background Model)을 사용하여 각 추적 영역들을 제한하였다. 또한 폐색영역에 가중치를 부가하여 확률분포영상에서 중심값을 이동시켜 폐색 영역을 회피하였다. 실험 결과 다중 물체들에 강인한 추적을 보이고 유사한 색상을 갖는 물체의 폐색 시에도 우수한 결과를 보임을 확인하였다.

• 한국컴퓨터정보학회논문지
• /
• 제15권12호
• /
• pp.27-35
• /
• 2010

유해영상 검출에서 정확하게 피부 색상 영역을 획득하는 것은 매우 중요하다. 그러나 기존의 방법들은 서로 다른 인종, 조명, 화장, 사용된 카메라 등과 같은 여러 원인으로 인해 피부 색상 추출에 여전히 문제를 가지고 있으며, 사전에 미리 정해진 피부 색상 분포 모델을 이용하여 영상에서 피부 영역을 검출한다. 이러한 문제를 해결하기 위해 본 논문에서는 눈 주변 영역에서 샘플을 추출하여 입력 영상에 최적으로 적합된 피부 색상 분포 모델을 생성하여 피부 영역을 강건하게 분할하고, 분할된 피부 영역에서 성인 영상을 대표할 수 있는 특징을 추출한 후, 분할된 피부 영역이 나체의 몸체를 포함하고 있는지를 뉴럴 네트워크 다층 퍼셉트론을 통해 여러 대표적인 특징들을 통합하면서 추론하는 새로운 방법을 제안하다. 본 논문의 실험에서는 피부 색상 영역 분할과 성인영상 검출의 두 가지 성능 측면에서 제안한 방법의 성능이 기존의 방법에 비해 보다 우수함을 보인다. 본 논문에서 제안한 강건한 유해영상검출 기법은 얼굴 검출, 성인영상 필터링 등과 같은 관련된 여러 응용 분야에서 유용하게 활용될 것이라 기대된다.

• 한국실내디자인학회논문집
• /
• 제16권5호
• /
• pp.22-29
• /
• 2007

In the face of the new dimension of the developed technology $L{\acute{a}}szl{\acute{o}}$ Moholy-Nagy, a master of 'the Bauhaus Akademie', made an effort to constitute the integrated life of the intellectual and the emotional by translating the experiences of the new technology Into the emotional language and the cultural realities. As an artist who criticized in the constructivist ethos the conception of the image as the imitation and brought the Image movements into relief Moholy-Nagy groped for a new expressive idea that can be called as 'the idea of the non-representational spatial expression' in which the expressive elements interact one another. His idea of the non-representational spatial expression that is made up of the pure forms of the nature, the direct colors and the spatial elements is oriented to incorporate the modern realities, i.e. the space experiences as the complicated sensory workings and the new technology as the measure of the human thinking, and further to see where they will move to. This study is on the one hand purposed to explicate the fundamental idea of the non-representational spatial expression that was poorly illuminated in spite of its importance in the respects of the design and art history, on the other to re-actualize the implications of the space-design which are contained in it.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.628-631
• /
• 2003

In this paper, we present real-time, accurate face region detection and tracking technique for an intelligent surveillance system. It is very important to obtain the high-resolution images, which enables accurate identification of an object-of-interest. Conventional surveillance or security systems, however, usually provide poor image quality because they use one or more fixed cameras and keep recording scenes without any cine. We implemented a real-time surveillance system that tracks a moving person using four pan-tilt-zoom (PTZ) cameras. While tracking, the region-of-interest (ROI) can be obtained by using a low-pass filter and background subtraction. Color information in the ROI is updated to extract features for optimal tracking and zooming. The experiment with real human faces showed highly acceptable results in the sense of both accuracy and computational efficiency.

• 대한한의진단학회지
• /
• 제10권2호
• /
• pp.121-131
• /
• 2006

Background : As each human has a look in the face of oneself, he or she has a look of him or herself in the shape of the body also. And for the shape of the body which gives a big clue in diagnosis in musculoskeletal disorders. Therefor many means are used and developed for diagnosis through body shape or posture analysis for musculoskeletal disorders. X-ray, CT, MRI has been used for diagnosis through image in this way to tell about the inside of the human body. On the other side, moire topography was used for information about the exterior of the human body, but yet only minimal information such as the number of contour lines in each side was available. Therefor there were a few studies to use moire topography or other methods to get information about the surface of the human body in numeric values. The instrument used in this study which is a laser projection moire, is another trial to get numeric data about the surface of the human body. The instrument is composed of laser projector and a computer software to recompose and analyse the image data into depth, height, angle and length. Objectives : The study was focused on whether the instrument is reliable for clinic use, and to seek the proper environment and posture for the examination, and among the data the software provides, which items are more reliable and useful. Methods : For reproductivity and repeatability, 4 testers tested 2 persons. And to how if the body shape changes according to the posture and which posture gives the most reliable data, the test was performed in 6 different positions. Results : Result, the instrument showed sufficient repeatability and reproductivity for clinical use. And among the items the software provides, the length of the back, the angle of the back in the sagittal and coronal plane showed reliable results. And there was difference in the results according to the posture, and Therefor, in following studies using this instrument or similar type of posture analysing instruments, the length of the back, the angle of the back in the sagittal and coronal plane could be reliable item to use.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 가을 학술발표논문집 Vol.30 No.2 (2)
• /
• pp.661-663
• /
• 2003

In this paper, we present new methods for rapid high-resolution range sensing using green-blue stripe pattern. We use green and blue for designing high-frequency stripe projection pattern. For accurate and reliable range recovery, we identify the stripe patterns by our color-stripe segmentation and unwrapping method. The experimental result for a naked human face shows the effectiveness of our method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권1호
• /
• pp.1-6
• /
• 2005

In this paper, we propose an emotion recognition method using the facial images and speech signals. Six basic emotions including happiness, sadness, anger, surprise, fear and dislike are investigated. Facia] expression recognition is performed by using the multi-resolution analysis based on the discrete wavelet. Here, we obtain the feature vectors through the ICA(Independent Component Analysis). On the other hand, the emotion recognition from the speech signal method has a structure of performing the recognition algorithm independently for each wavelet subband and the final recognition is obtained from the multi-decision making scheme. After merging the facial and speech emotion recognition results, we obtained better performance than previous ones.