• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.1913_1914
• /
• 2009

본 논문에서 조명환경하에서 음성/영상 연동시스템을 통해서 외부음향잡음의 차단을 위한 강인한 입술움직임 영상영역을 추적하는 한 가지 방법을 제안한다. 조명환경하에서 강인한 입술움직임 영상영역을 추적하기 위해 온라인상에서 입술움직임 표준영상을 수집하였고 다양한 조명환경에 적응하는 입술 움직임 영상의 특징들을 추출하였다. 동시에 온라인 템플릿 영상을 획득하였고, 이 영상들을 템플릿 정합을 위해 사용했다. 음성/영상처리시스템의 연동결과, 다양한 조명환경하에서 그 연동률을 99.3%까지 높일 수 있었고 음향잡음에 의한 음성인식 실행을 원천적으로 차단할 수 있었다.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.11
• /
• pp.1472-1475
• /
• 1999

This paper proposes a new detection method of human faces in grey scale images with cluttered background using a facial template and elliptical structure of the human head. Face detection technique can be applied in many areas of image processing such as face recognition, composition and computer graphics, etc. Until now, many researches about face detection have been done, and applications in more complicated conditions are increasing. The existing technique proposed by Sirohey shows relatively good performance in image with cluttered background, but can apply only to image with one face and needs much computation time. The proposed method is designed to reduce complexity and be applied even in the image with several faces by introducing template matching as preprocess. The results show that the proposed method produces more correct detection rate and needs less computation time than the existing one.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.23 no.3
• /
• pp.233-238
• /
• 2005

3-D virtual city model is becoming increasingly important for a number of GIS applications. For reconstruction of 3D building in urban area aerial images, satellite images, LIDAR data have been used mainly and most of researches related to 3-D reconstruction focus on development of method for extraction of building height and reconstruction of building. In case of automatically extracting and reconstructing of building height using only aerial images or satellite images, there are a lot of problems, such as mismatching that result from a geometric distortion of optical images. Therefore, researches of integrating optical images and existing digital map (1/1,000) has been in progress. In this paper, we focused on extracting of building height by means of interest points and vertical line locus method for reducing matching points. Also we used digital plotter in order to validate for the results in this study using aerial images (1/5,000) and existing digital map (1/1,000).

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.335-336
• /
• 2007

본 논문은 표면실장기판을 위한 자동 광학 검사 시스템에서 사용되는 알고리즘을 종류 별로 비교 분석한다. 비교된 부품 검사 알고리즘은 공간영역에서의 템플릿 매칭 방법을 사용한다. 각 방법에는 장단점이 있으며 주된 비교항목은 검사 속도와 검사 정확도에 대한 비교 분석을 한다. 비교 및 분석을 통해 부품검사 알고리즘의 속도 및 정확도 향상의 방향을 제시한다. 실험 결과를 제시하여 비교된 알고리즘의 성능을 검증한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.443-444
• /
• 2020

본 논문에서는 여러 개의 측방 유동 스트립을 형광분석을 통한 정량분석을 할 수 있는 장비의 각 스트립에 인쇄된 바코드를 카메라를 이용하여 인식하는 방법을 제안한다. 제안한 알고리즘에서는 각 슬롯의 스트립 유무를 판단하고, 시작 비트의 위치를 템플릿 정합법으로 검출하여 바코드 영역을 찾는다. 각 비트 영역은 바코드 설계 데이터와 기기 교정 시 계산된 공간 해상도를 이용하여 결정된다. 각 비트의 값은 비트 영역 중앙 부분의 평균을 이용하여 결정하였다. 다양한 조명 아래에서 취득한 영상들로부터 스트립 유무 판단, 시작 비트 위치 탐색 성공 여부 및 각 비트 값을 결정 등을 위한 판정 값을 가우시안 모델을 이용하여 계산하였다. 실험 결과 모든 판정 오류는 무시할 만 하였다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.1
• /
• pp.7-12
• /
• 2011

In this paper, we propose the technique detecting interest object region effectively in the images from periscope of submarine based on self-generated template. First, we extract the sea-sky line, and divide it into sky and sea area from background region based on the sea-sky line. In each divided background region, the blocks which can be represented in each background region are set as a reference template. After dividing an image into several same size of blocks, we apply multi template matching to the divided search blocks and histogram template to divide the image into object region and background region. Proposed algorithm is adapted to various images in which objects exist in the background of sea and sky. We verified that proposed algorithm performed properly without given informmed prby prior learning.ropso, regardless of the slope of sea-sky line and the locmed p of object based on sea-sky line, we verified that the objects region was segmented effectively from the input image.

• Journal of Korea Multimedia Society
• /
• v.15 no.6
• /
• pp.722-730
• /
• 2012

Recently, eye image information has been widely used for iris recognition or gaze detection in biometrics or human computer interaction. According as long distance camera-based system is increasing for user's convenience, the noises such as eyebrow, forehead and skin areas which can degrade the accuracy of eye detection are included in the captured image. And fast processing speed is also required in this system in addition to the high accuracy of eye detection. So, we compared the most widely used algorithms for eye detection such as AdaBoost eye detection algorithm, adaptive template matching+AdaBoost algorithm, CAMShift+AdaBoost algorithm and rapid eye detection method. And these methods were compared with images including light changes, naive eye and the cases wearing contact lens or eyeglasses in terms of accuracy and processing speed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.2
• /
• pp.308-314
• /
• 2006

A line scan camera is applied to enhance COG(Chip On Glass) inspection accuracy to be measured a few micro unit. The foreign substance detection among various faulty factors has been the most difficult technology in the faulty automatic inspection step since COG pattern is very miniature and complexity. In this paper, we proposed two step area segmentation template matching method to increase matching speed. Futhermore to detect foreign substance(such as dust, scratch) with a few micro unit, the new method using gradient mask and AND operation was proposed. The proposed 2 step template matching method increased 0.3 - 0.4 second matching speed compared with conventional correlation coefficient. Also, the proposed foreign substance applied masks enhanced $5-8\%$ faulty detection rate compared with conventional no mask application method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.602-607
• /
• 2002

In this paper, we proposed an approach to efficiently compress and transmit multiresoltion 3D lariat models for multimedia and very low bit rate applications. A personal facial model is obtained by a 3D laser digitizer, and further re-quantized at several resolutions according to different scope of applications, such as animation, video game, and video conference. By deforming 2D templates to match and re-quantize a 3D digitized facial model, we obtain its compressed model. In the present study, we create hierarchical 2D lariat wireframe templates are adapted according to facial feature points and the proposed piecewise chainlet affined transformation(PACT) method. The 3D digitized model after requantization are reduced significantly without perceptual loss. Moreover the proposed multiresoulation lariat models possessed of hierarchial data structure are apt to be progressively transmitted and displayed across internet.

• Korean Journal of Cognitive Science
• /
• v.14 no.1
• /
• pp.11-16
• /
• 2003

This Paper extracts the edge of main components of face with Gator wavelets transformation in facial expression images. FCM(Fuzzy C-Means) clustering algorithm then extracts the representative feature points of low dimensionality from the edge extracted in neutral face. The feature-points of the neutral face is used as a template to extract the feature-points of facial expression images. To match point to Point feature points on an expression face against each feature point on a neutral face, it consists of two steps using a dynamic linking model, which are called the coarse mapping and the fine mapping. This paper presents an automatic extraction of feature-points by dynamic linking model based on Gabor wavelets and fuzzy C-means(FCM) algorithm. The result of this study was applied to extract features automatically in facial expression recognition based on dimension[1].