• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.337-342
• /
• 2008

최근 들어 기존의 녹색 바탕 차량 번호판에서, 흰색 바탕의 신 차량 번호판으로 교체되고 있다. 하지만 아직 기존 차량 번호판이 신 차량 번호판으로 전면 교체되지 않아 두 번호판 모두 사용되고 있다. 따라서 주차관리 시스템, 속도위반, 신호 위반 등 무인 카메라를 이용한 시스템에서, 기존 차량 번호판과 신 차량 번호판의 특징에 맞는 인식 시스템이 요구된다. 본 논문에서는 이러한 문제를 해결하기 위해 기존 차량 번호판과 신 차량 번호판을 통합한, 지능형 차량 번호판 인식 시스템을 제안한다. 무인 카메라에서 획득된 차량 영상에서 번호판의 색상 정보를 이용하여 기존 차량 번호판과 신 차량 번호판을 구분한다. 기존 차량 번호판인 경우에는 HSI 컬러 공간을 이용하여 이진화를 적용하며, 신 차량 번호판인 경우에는 블록 이진화를 적용한다. 이진화된 영상을 대상으로 차량의 형태학적 특징을 이용하여 잡음을 제거한 후, 차량 번호판 영역을 추출한다. 추출된 차량 번호판 영역에 대해 Labeling 알고리즘을 적용하여 개별 문자를 추출한다. 추출된 개별 문자는 FCM 알고리즘을 적용하여 인식한다. 제안된 차량 번호판 추출 및 인식 방법의 성능을 평가하기 위해 160장의 기존 차량 영상과 100장의 신 차량 영상을 대상으로 실험한 결과, 제안된 차량 번호판 추출 및 인식 방법이 실험을 통해서 효율적인 것을 확인하였다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.90-95
• /
• 2012

Remotely sensed data provide valuable information on land monitoring due to multi-temporal observation over large areas. Especially, high resolution imagery with 0.6~1.0 m spatial resolutions contain a wealth of information and therefore are very useful for thematic mapping and monitoring change in urban areas. Recently, remote sensing technology has been successfully utilized for natural disaster monitoring such as forest fire, earthquake, and floods. In this paper, an efficient change detection method based on texture differences observed from high resolution multi-temporal data sets is proposed for mapping disaster damage and extracting damage information. It is composed of two parts: feature extraction and detection process. Timely and accurate information on disaster damage can provide an effective decision making and response related to damage.

• Korean Journal of Remote Sensing
• /
• v.32 no.6
• /
• pp.693-701
• /
• 2016

To make a mosaicked image, we need to estimate the geometric relationship between individual images. For such estimation, we needs tiepoint information. In general, feature-based methods are used to extract tiepoints. However, in the case of textureless images, feature-based methods are hardly applicable. In this paper, we propose a frequency domain matching method for automated mosaicking of textureless images. There are three steps in the proposed method. The first step is to convert color images to grayscale images, remove noise, and extract edges. The second step is to define a Region Of Interest (ROI). The third step is to perform phase correlation between two images and select the point with best correlation as tiepoints. For experiments, we used GOCI image slots and general frame camera images. After the three steps, we produced reliable tiepoints from textureless as well as textured images. We have proved application possibility of the proposed method.

• Journal of Korea Multimedia Society
• /
• v.7 no.5
• /
• pp.648-657
• /
• 2004

A method of extracting central objects in color images without any prior-knowledge is proposed in this paper, which uses basically information of significant color distribution. A central object in an image is defined as a set of regions that lie around center of the image and have significant color distribution against the other surround (or background) regions. Significant colors in an image are first defined as the colors that are distributed more densely around center of the image than near borders. Then core object regions (CORs) are selected as the regions a lot of pixels of which have the significant colors. Finally, the adjacent regions to the CORs are iteratively merged if they are similar to the CORs but not to the background regions in color distribution. The merging result is accepted as the central object that may include differently color-characterized regions and/or two or more objects of interest. Usefulness of the significant colors in extracting the central object was verified through experiments on several kinds of test images. We expect that central objects shall be used usefully in image retrieval applications.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.7
• /
• pp.655-661
• /
• 2017

It is a very important task to extract character regions contained in various input color images, because characters can provide significant information representing the content of an image. In this paper, we propose a new method for extracting character regions from various input images using MCT features and an AdaBoost algorithm. Using geometric features, the method extracts actual character regions by filtering out non-character regions from among candidate regions. Experimental results show that the suggested algorithm accurately extracts character regions from input images. We expect the suggested algorithm will be useful in multimedia and image processing-related applications, such as store signboard detection and car license plate recognition.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.05a
• /
• pp.437-442
• /
• 2005

본 연구에서는 칼라정보와 주성분분석법(principal component analysis : PCA)를 이용한 차량 번호판 인식시스템을 구성하였다. 먼저 입력된 차량 영상에서 번호판의 형태적 특징과 녹색 칼라 정보를 이용하여 번호판 영역을 추출하였으며, 추출된 번호판내의 문자 및 숫자의 위치적 특징을 이용하여 번호판의 종류(구형, 신형, 최신형)를 구분하였다. 이렇게 추출되고 구분된 번호판은 문자의 상대적 위치정보와 수평 및 수직 투영 정보를 함께 이용하여 각각의 문자영역을 분리 추출하였다. 추출된 문자영역은 주성분분석법을 이용하여 고유벡터를 추출한 후 문자 인식에 사용하였다. 본 논문의 실험과정에서는 다양한 시간대 환경에서 촬영된 주행 중인 자동차 320대의 자가용 차량영상에 대하여 실험하였으며 높은 번호판 추출률과 번호판종류 구분률 그리고 문자 인식률을 얻을 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.118-121
• /
• 2003

지문을 이용한 개인 인증 절차는 지문 형태 별로 구분하는 분류(classification) 과정과 본인임을 확인하는 정합(matching) 과정으로 구분할 수 있다. 지문의 분류와 정합을 위해서는 기존 연구들이 지문의 특징점 수와 방향성의 흐름 패턴에 의존한다. 본 논문에서는 방향성의 흐름 패턴을 이용한 중심점 추출에 초점이 맞춰져 있으며 추출된 중심점 정보는 현재 구현되어진 특징점 추출 정보와 연계해 정합을 위한 기준점으로 활용한다. 중심점 추출 방식은 입력된 지문 영상에 대해 3 $\times$ 3 Sobel 마스크를 적용한 후 8 $\times$ 8블록 영상을 분할하여 각 대표 방향 성분을 추출하며 추출되어진 방향 성분과 특이점 패턴을 비교하여 중심점을 탐색한다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.2
• /
• pp.47-54
• /
• 2001

The main purpose of this study is to propose the algorithm about the extraction of the facial feature. To achieve the above goal, first of all, this study produces binary image for input color image. It calculates area after pixel labeling by variant block-units. Secondly, by contour following, circumference have been calculated. So the proper degree of resemblance about area, circumference, the proper degree of a circle and shape have been calculated using the value of area and circumference. And Third, the algorithm about the methods of extracting parameters which are about the feature of eyes, nose, and mouse using the proper degree of resemblance, general structures and characteristics(symmetrical distance) in face have been accomplished. And then the feature parameters of the front face have been extracted. In this study, twelve facial feature parameters have been extracted by 297 test images taken from 100 people, and 92.93 % of the extracting rate has been shown.

• Journal of the Korea Society of Computer and Information
• /
• v.9 no.2
• /
• pp.19-26
• /
• 2004

This paper proposes ACA (Adaptive Component Analysis) as a method for feature extraction and analysis of the content-based image retrieval system. For satisfactory retrieval, the features extracted from images should be appropriately applied according to the image domains and for this, retrieval measurement is proposed in this study. Retrieval measurement is a standard indicating how important the value of a relevant feature is to image retrieval. ACA is a middle stage for content-based image retrieval and it purposes to improve the retrieval speed and performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.113-117
• /
• 2003

In this paper, have necessity of PACS (Picture Archiving and Communication System) at hospital but hereafter by economical problem PACS apply this to medical treatment image enhancing image quality applying histogram equalization for improvement of light and darkness after reconstruct because make image that pretreatment filtering has wild picture and is processed in wave lets dissolution and wave lets area using weight median filter because could not buy expensive equipment at hospital which introduction is difficulty do inversion and extracted characteristic.