• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.48 no.4
• /
• pp.6-13
• /
• 2011

Due to its robustness to illumination change, normalized cross-correlation based similarity measurement is widely used in many machine vision applications. However, its inefficient computation structure is not adequate for real-time embedded vision system. In this paper, we present an efficient hardware architecture based on a normalized cross correlation (NCC) for fast image similarity measure. The proposed architecture simplifies window-sum process of the NCC using the integral-image. Relieving the overhead to constructing integral image, we make it possible to process integral image construction at the same time that pixel sequences are inputted. Also the proposed segmented integral image method can reduce the buffer size for storing integral image data.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.580-582
• /
• 2000

본 논문에서는 MAM을 이용한 특징 벡터의 보정을 기반으로 하는 헤드 제스처 인식에 관해 기술한다. 제안된 시스템은 얼굴 움직임 검출 모듈과 눈 영역 추적 모듈, 미 측정된 벡터 보정 모듈, 측정된 제스처에 대한 인식모듈로 구성된다. 신경망과 모자이크 이미지를 이용하여 얼굴 영역을 검출하고, 이 영역에서 눈 영역을 검출한다. 만약 눈의 쌍이 검출되지 않는다면 시스템은 특징 벡터 보정(MAM)을 수행하여 손실된 정보를 예측한다. 검출된 눈 영역은 정규화된 벡터로 변경된다. 이 벡터의 분산을 이용하여 긍정, 부정, 중립의 제스처를 판단한다. 제스처의 인식은 직접 관측, 이중 HMM, 삼중 HMM을 사용한 다중 인식기를 이용한다.

• Proceedings of the KSRS Conference
• /
• 2009.03a
• /
• pp.101-104
• /
• 2009

기후 시스템에서 지구온난화는 세계적으로 매우 중요한 문제이고 이는 기후변화, 이상기온, 폭우, 가뭄 등의 문제를 초래한다. 특히 사막화는 전 세계적으로 10억 명 이상의 사람들에게 영향을 미치고 있다. 건조한 상태의 식생은 사막화되기 쉽기 때문에 식생의 수분상태는 사막화의 중요한 지표이다. 본 논문에서는 중국과 몽골 사막 주변영역에 대해 식생의 수분상태를 탐지하였다. 식생의 수분상태를 탐지하기 위해 1999년부터 2006년까지의 SPOT/VEGETATION 위성 이미지를 이용하여 정규화 수분지수(NDWI: Normalized Difference Water Index)를 산출하였다. 그 결과 1999년부터 2006년까지의 NDWI는 사막주변영역에서 감소하는 경향을 보였고, 그 영역은 몽골 고비사막 북동지역과 중국 타클라마칸 사막의 남동지역에 위치해 있었다.

• The Journal of the Korea Contents Association
• /
• v.11 no.5
• /
• pp.125-137
• /
• 2011

Recently the research on the system using images taken from camera phones as input is actively conducted. This paper proposed a system that shows wine pictures which are similar to the input wine label in order. For the calculation of the similarity of images, the representative color of each cell of the image, the recognized text color, background color and distribution of feature points are used as the features. In order to calculate the difference of the colors, RGB is converted into CIE-Lab and the feature points are extracted by using Harris Corner Detection Algorithm. The weights of representative color of each cell of image, text color and background color are applied. The image similarity is calculated by normalizing the difference of color similarity and distribution of feature points. After calculating the similarity between the input image and the images in the database, the images in Database are shown in the descent order of the similarity so that the effort of users to search for similar wine labels again from the searched result is reduced.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.571-573
• /
• 2000

복잡한 이미지에서 얼굴 추출은 얼굴 영상처리 분야에서 기본적이면서도 배경이 복잡함으로 인해 많은 어려움이 따른다. 이 논문에서는 복잡한 화면 이미지에서 얼굴을 추출하기 위해 여러 가지 과정을 거친다. 다양한 피부색을 가진 얼굴에 대해 즉, 흑인과 황인, 백인 등을 모두 추출하기 위해 피부색 모델을 이용한다. 다양한 피부색에 대한 임계값(threshold)을 이용하여 피부색과 다른 영역을 구분하여 얼굴의 후보 데이터로 추출한다. 그 추출된 후보 데이터를 지역적 임계값(local threshold)을 이용하여 얼굴과 눈, 코, 입과 같은 세부사항에 분류한다. 분류된 부분이 즉 얼굴내에서 얼굴이 아닌 부분(눈, 코, 입 등)의 크기가 정규화 되어진 최소 크기보다 박을 경우 그 후보 데이터를 버리고, 그렇지 않을 경우, 즉 얼굴이 아닌 다른 부분의 크기가 정해진 크기보다 크거나 같을 경우 그 후보 데이터를 검출한다. 이 논문에 결과는 배경에서도 피부색과 얼굴의 부분영역을 이용하여 얼굴을 검출할 수 있다는 것을 보인다.

• Journal of Game and Entertainment
• /
• v.2 no.2
• /
• pp.45-50
• /
• 2006

In a face detection system based on template matching, histogram equalization or log transform is applied to an input image for the intensity normalization and the image improvement. It is known that they are noneffective in improving an image with intensity distortion by illumination variation. In this paper, we propose an efficient image improvement method called as a bilateral symmetry average for images with intensity distortion by illumination variation. Experimental results show that our method delivers the detection performance better than previous methods and also remarkably reduces the number of face candidates.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.6
• /
• pp.903-917
• /
• 1995

In most database systems images have been indexed indirectly using related texts such as captions, annotations and image attributes. But there has been an increasing requirement for the image database system supporting the storage and retrieval of images directly by content using the information contained in the images. There has been a few indexing methods based on contents. Among them, Pertains proposed an image indexing method considering spatial relationships and properties of objects forming the images. This is the expansion of the other studies based on '2-D string. But this method needs too much storage space and lacks flexibility. In this paper, we propose a more flexible index structure based on kd-tree using paging techniques. We show an example of extracting keys using normalization from the from the raw image. Simulation results show that our method improves in flexibility and needs much less storage space.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2006.05a
• /
• pp.264-267
• /
• 2006

One of the serious problems that make the flaw identification in a multi-layered thick composite panel more difficult is the interferenceeffect of the upper layer. To take care of such a problem, here we propose an image enhancement approach that can get rid of such an interference effect to ultrasonic C-scan images by a normalization of the acquired signals by a reference signals, and demonstrate its performance in the experiments. Specifically, three specimens with artificial flaws are prepared and ultrasonic C-scan images are acquired experimentally to eliminate the undesired interference effect. Cleat successes are observed in the present study demonstrating the high potential of the proposed algorithm as a practical image enhancement tool in many practical situations.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.165-170
• /
• 2003

This paper presents an integrated approach to unconstrained face recognition in arbitrary scenes. The front end of the system comprises of a scale and pose tolerant face detector. Scale normalization is achieved through novel combination of a skin color segmentation and log-polar mapping procedure. Principal component analysis is used with the multi-view approach proposed in[10] to handle the pose variations. For a given color input image, the detector encloses a face in a complex scene within a circular boundary and indicates the position of the nose. Next, for recognition, a radial grid mapping centered on the nose yields a feature vector within the circular boundary. As the width of the color segmented region provides an estimated size for the face, the extracted feature vector is scale normalized by the estimated size. The feature vector is input to a trained neural network classifier for face identification. The system was evaluated using a database of 20 person's faces with varying scale and pose obtained on different complex backgrounds. The performance of the face recognizer was also quite good except for sensitivity to small scale face images. The integrated system achieved average recognition rates of 87% to 92%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.6
• /
• pp.1312-1317
• /
• 2003

This paper presents an integrated approach to unconstrained face recognition in arbitrary scenes. The front end of the system comprises of a scale and pose tolerant face detector. Scale normalization is achieved through novel combination of a skin color segmentation and log-polar mapping procedure. Principal component analysis is used with the multi-view approach proposed in[10] to handle the pose variations. For a given color input image, the detector encloses a face in a complex scene within a circular boundary and indicates the position of the nose. Next, for recognition, a radial grid mapping centered on the nose yields a feature vector within the circular boundary. As the width of the color segmented region provides an estimated size for the face, the extracted feature vector is scale normalized by the estimated size. The feature vector is input to a trained neural network classifier for face identification. The system was evaluated using a database of 20 person's faces with varying scale and pose obtained on different complex backgrounds. The performance of the face recognizer was also quite good except for sensitivity to small scale face images. The integrated system achieved average recognition rates of 87% to 92%.