• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -1
• /
• pp.441-445
• /
• 2000

This study is to segment white matter, gray matter, and cerebrospinal fluid(CSF) on a brain MR image of coronal section and to calculate the volume of each. First, we segmented the whole region of a brain from a black colored background, a skull and a fat layer. Then, we calculated the partial volume of each component, which was present in scanning finite thickness, with the arithmetical analysis of gray value from the internal region of a brain showing the blurring effects on the basis of the MR image forming principle. Calculated partial volumes of white matter, gray matter and CSF were used to determine the threshold for the segmentation of each component on a brain MR image showing the blurring effects. Finally, the volumes of segmented white matter, gray matter, and CSF were calculated. The result of this study can be used as the objective diagnostic method to determine the degree of brain atrophy of patients who have neurodegenertive diseases such as Alzheimer’s disease and cerebral palsy.

• 제어로봇시스템학회논문지
• /
• 제15권9호
• /
• pp.939-945
• /
• 2009

In this paper, we proposed a method which corresponds people under the structured spaces with multiple cameras. The correspondence takes an important role for using multiple camera system. For solving this correspondence, the proposed method consists of three main steps. Firstly, moving objects are detected by background subtraction using a multiple background model. The temporal difference is simultaneously used to reduce a noise in the temporal change. When more than two people are detected, those detected regions are divided into each label to represent an individual person. Secondly, the detected region is segmented as features for correspondence by a criterion with the color distribution and context information of human body. The segmented region is represented as a set of blobs. Each blob is described as Gaussian probability distribution, i.e., a person model is generated from the blobs as a Gaussian Mixture Model (GMM). Finally, a GMM of each person from a camera is matched with the model of other people from different cameras by maximum likelihood. From those results, we identify a same person in different view. The experiment was performed according to three scenarios and verified the performance in qualitative and quantitative results.

• 한국통신학회논문지
• /
• 제24권10B호
• /
• pp.1986-1994
• /
• 1999

본 논문에서는, 동영상에서 의미 있는 객체 영역을 추출하기 위해서, 첫 장의 영상 분할은 사람에 의해서 주어진 것으로 가정하고, 그 다음 프레임부터는 사람의 도움 없이 객체를 추적해 가는 반자동 방식의 이동체 추적 알고리즘을 제안한다. 제안하는 알고리즘은 계층적인 구조를 가지며, 각각의 계층에서는 seed 추출, watershed 알고리즘을 이용한 영상 분할, 영역 구분의 단계를 거쳐 영상 분할을 수행한다. 영역 구분 단계에서는, 순방향으로 추정된 움직임 벡터장으로부터 영역 분할의 판단 기준을 만들고 이를 이용하여 각각의 영역을 '객체 영역', '배경 영역', '불확실 영역'으로 구분한다. 이때, '불확실 영역'으로 구분된 영역들에 대해서는 좀 더 낮은 계층에서 위의 단계들을 반복하여 다시 수행하게 한다. 제안한 알고리즘은 컴퓨터 모의실험을 통해서 'Claire', 'Miss America', 'Akiyo', 'Mother and daughter'의 영상에서 바람직한 추적 결과를 나타냄을 확인하였다.

• 한국콘텐츠학회논문지
• /
• 제7권2호
• /
• pp.40-50
• /
• 2007

본 논문에서는 카메라로부터 획득 되어진 비디오 시퀀스로부터 다중 움직임 객체와 배경을 분할하고 시공간 정보에 기반 한 객체 추적 방법을 제안한다. 제안한 방법은 3단계로 구성되어 있다. 먼저 입력 비디오 시퀀스로부터 프레임 사이의 차를 이용한 움직임 영역과 움직임이 존재하지 않는 영역을 구분하여 적응적 경계간을 추출한다. 두 번째는 참조 배경영상과 적응적 경계값을 이용하여 움직임이 존재하는 영역으로부터 개략적 객체 분할을 수행하며, 분할된 이진영상에 형태학적 영역 병합 알고리즘을 적용하여 객체 병합을 수행하였다. 마지막으로 분할된 객체에 시공간 정보를 이용하여 객체에 임의의 ID를 할당하여 추적하였다. 카메라로부터 획득되어진 비디오 시퀀스를 이용한 실험에서 객체들의 분할 및 추적의 효율성과 시스템의 유용성을 확인하였다.

• 한국한의학연구원논문집
• /
• 제14권3호
• /
• pp.89-95
• /
• 2008

A tongue shows physiological and clinicopathological changes of inner organs. Visual inspection of a tongue is not only convenient but also non-invasive. To develop an automat ic tongue diagnosis system for an objective and standardized diagnosis, the separation of the tongue are a from a facial image and the detection of coatings, spots and cracks are inevitable but difficult since the colors of a tongue, lips, and skin in a mouth as well as those of tongue furs and body are similar. The propose d method includes preprocessing with down-sampling and edge enhancement, over-segmentation, detecting positions with a local minimum over shading from the structure of a tongue, and correcting local minima or detecting edge with color difference. The proposed method produces the region of a segmented tongue, and then decomposes the color components of the region into hue, saturation and brightness, resulting in classifying the regions of tongue furs(coatings) into kinds of coatings and substance and segmenting them. Spots are detected by using local maxima and the variation of saturation, and cracks are searched by using local minima and the directivity of dark areas in brightness. The results illustrate the segmented region with effective information, excluding a non-tongue region and also give us accurate discrimination of coatings and the precise detection of spots and cracks. It can be used to make an objective and standardized diagnosis for an u-Healthcare system as well as a home care system.

• 제어로봇시스템학회논문지
• /
• 제8권4호
• /
• pp.327-332
• /
• 2002

The main problems of computational complexity in object tracking are definition of objects, segmentations and identifications in non-structured environments with erratic movements and collisions of objects. The object's information as a region that corresponds to objects without discriminating among objects are considered. This paper describes the algorithm that, automatically and efficiently, recognizes and keeps tracks of interest-regions selected by users in video or camera image sequences. The block-based feature matching method is used for the region tracking. This matching process considers only dominant feature points such as corners and curved-edges without requiring a pre-defined model of objects. Experimental results show that the proposed method provides above 96% precision for correct region matching and real-time process even when the objects undergo scaling and 3-dimen-sional movements In successive image sequences.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1998년도 Proceedings of International Workshop on Advanced Image Technology
• /
• pp.29-33
• /
• 1998

In this paper, we propose a hierarchical segmentation method for tracking a semantic video object using a watershed algorithm based on morphological filtering. In the proposed method, each hierarchy consists of three steps: First, markers are extracted on the simplified current frame. Second, region growing by a modified watershed algorithm is performed for segmentation. Finally, the segmented regions are classified into 3 categories, i.e., inside, outside, and uncertain regions according to region probability values, which are acquired by the probability map calculated from a estimated motion field. Then, for the remaining uncertain regions, the above three steps are repeated at lower hierarchies with less simplified frames until every region is decided to a certain region. The proposed algorithm provides prospective results in video sequences such as Miss America, Clair, and Akiyo.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2002년도 추계종합학술대회
• /
• pp.240-243
• /
• 2002

본 논문에서는 칼라 이미지에서 색상 요소를 기초로 하여 얼굴영역을 추출하고 얼굴의 특징요소를 추출하는 방법을 제안한다. 얼굴 영역을 추출하기 위하여 일반적인 얼굴색상분포를 이용하여 색상변환을 하였다. 얼굴 특성요소를 찾기 위하여 윤곽선검출을 이용하였다. 얼굴영역의 상단부분에서 눈의 요소를 찾고, 눈과 입의 지정학적 위치를 이용하여 입의 후보영역을 지정하고 입을 찾도록 하였다. 검색영역을 좁혀 계산량을 줄임으로서 탐색시간을 줄일 수 있고, 일반적인 얼굴색상분포를 이용하여 얼굴 영역을 검출함으로서 얼굴표정, 얼굴색변화, 기울짐에 대해서도 얼굴영역을 검출할 수 있었다.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제10권2호
• /
• pp.67-80
• /
• 2006

본 논문에서는 스테레오 비전을 이용하여 기존의 포인팅 디바이스인 마우스를 대신할 새로운 포인팅 디바이스 방법을 제안 한다. 제안한 방법은 기존의 인식 장비들이 가진 마커에 의한 움직임 제약과 고가 장비의 단점을 극복하기 위해 컴퓨터 비전을 이용한다. 즉, 사람이 시각을 통해 정보를 인식하는 것과 동일하게 컴퓨터의 영상 정보를 이용하여 실시간으로 영상내의 컬러 영역의 분할을 통해 물체의 위치를 추적 및 정합하고 이의 위치를 스테레오 기하학 정보로부터 계산하여 포인팅 동작을 수행한다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제8권1호
• /
• pp.21-28
• /
• 2002

This paper presents a brush stroke direction creation method to distill painterly expressions from picture images. In actual paintings, each brush stroke has the same directionality in one segmented area. but directionality changes in the area of image border edges. This study employed the following methods to implement the directionality of the brush stroke area by: (1) detecting edge in input images, (2) region splitting and merging based on quad-tree division method, and (3) composing a direction map to create brush strokes with the same directionality in the segmented area. The results of the implementation enabled the simulation of the directionality of the stroke area shown in actual paintings.