• Journal of the Korean Data and Information Science Society
• /
• v.21 no.4
• /
• pp.689-697
• /
• 2010

In this study, we propose a new procedure for improving the character recognition of text area extracted from video images. The recognition of strings extracted from video, which are mixed with Hangul, English, numbers and special characters, etc., is more difficult than general character recognition because of various fonts and size, graphic forms of letters tilted image, disconnection, miscellaneous videos, tangency, characters of low definition, etc. We improved the recognition rate by taking commonly used letters and leaving out the barely used ones instead of recognizing all of the letters, and then using SVM and RDA character recognition methods. Our numerical results indicate that combining SVM and RDA performs better than other methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2005.11a
• /
• pp.275-278
• /
• 2005

본 연구는 디스플레이 제품의 개발 및 생산과정에서 OSD 메뉴문자의 오류 유무를 검사하는 과정을 컴퓨터 비전기술을 사용하여 자동화하는 방법을 제안한다. 디스플레이 제품의 OSD 메뉴는 순차적인 제어과정을 통해서 제한된 디스플레이 영역에 여러 종류의 언어와 기호를 포함하는 형태로 출력된다. 기존의 제품개발 과정에서 이러한 메뉴 항목의 정확성을 검증하는 작업은 작업자의 육안에 의한 판단과 수작업에 의해 이루어지고 있는데, 이는 반복작업에 의한 집중력 저하 및 판단착오에 의한 오류의 가능성을 내재한다. 또한 작업자가 다양한 나라의 언어에 대한 문자형태와 기호표현의 특성을 이해하여야 하고, 검증작업 자체에 따르는 부수적인 시간과 노력을 필요로 한다. 이에 본 연구에서는 디스플레이 제품의 OSD 메뉴와 같이 특수한 구조를 갖는 문서영상에 대한 논리적인 구조분석을 통해서 연속적인 문서영상을 발생시키는 작업스케쥴러를 생성하고, 작업스케쥴러에 의해 순차적으로 발생된 영상문서에 대한 전처리, OSD 메뉴의 기하학적 구조분석 및 문자영역을 추출하는 방법과, 표준패턴 구축 및 원형정합에 의한 문자의 오류를 검증하는 방법과 오류를 관리하는 기법을 제안한다.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.12
• /
• pp.51-57
• /
• 1999

In this paper, an adaptive thresholding of the nonuniformly contrasted images is proposed through using the contrast pre-enhancement of the local regions and the bilinear interpolation between the local threshold values. The nonuniformly contrasted image is decomposed into 9${\times}$9 sized local regions, and the contrast is enhanced by intensifying the gray level difference of each low contrasted or blurred region. Optimal threshold values are obtained by iterative method from the gray level distribution of each contrast-enhanced local region. Discontinuities are reduced at the region of interest or at the characters by using bilinear interpolation between the neighboring threshold surfaces. Character recognition experiments are conducted using backpropagation neural network on the characters extracted from the nonuniformly contrasted document, PCB, and wafer images binarized through using the proposed thresholding and the conventional thresholding methods, and the results prove the relative effectiveness of the proposed scheme.

• The Journal of the Convergence on Culture Technology
• /
• v.4 no.4
• /
• pp.361-366
• /
• 2018

For each character constituting learning data, different fonts are added in pixel unit to create MASK, and then pixel values belonging to the MASK are divided into three groups. The experimental data are modified into skeletal forms, and then regional boundary operation is used to create a boundary that distinguishes the background region adjacent to the skeleton of the character from the background of the modified experimental data. Discordance values between the modified experimental data and the MASKs are calculated, and then the MASK with the minimum value is found. This MASK is selected as a finally recognized result for the given experiment data. The recognition algorithm using skeleton of the character and the regional boundary operation can easily extend the learning data set by adding new fonts to the given learning data, and also it is simple to implement, and high character recognition rate can be obtained.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.11 no.1
• /
• pp.103-108
• /
• 2011

The modern society with the wide spread recognition of the importance of informatics and for the development of information and communication technology is rapidly processing. Especially the rapid development in mobile technology boost up the general expectation that one can get the information he wants anytime anywhere. Accordingly image search for the convenient information retrieval is becoming common. However general image search has difficulties because inexactitude extracting character in the image and getting the detail information in extracted character. Therefore these paper make character recognition through the images that I photographed a sightseeing resort, a signboard of a lot of stores to a smart phone camera, so information offer to be convenient to users is a purpose. A user can get detailed information, by character extraction way called top-hat algorithm and connect to a server.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.8A
• /
• pp.805-813
• /
• 2002

Document image should be classified into detailed regions as text, picture, table and etc through the geometric layout analysis if paper documents can be converted automatically into electronic documents. However, complexity of the document layout and variety of the size and density of a picture are the reason to make it difficult to analyze the geometric layout of the document images. In this paper, we propose the method which have a better performance of the region segmentation and classifications, and the line extraction in the table region than the commercial softwares and previous methods. The proposed method can segment the document into detailed regions by using connected components method even if its layout is complex. This method also classifies texts and pictures by using separable median filter even. Though their size and density are diverse, In addition, this method extracts the lines from the table adapting one dimensional median filter to the each horizontal and vertical direction, even though lines are deformed or texts attached to them.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2004.11a
• /
• pp.279-286
• /
• 2004

본 논문에서는 출입국자 관리의 효율성과 체계적인 출입국 관리를 위하여 위조 여권을 판별할 수 있는 지능형 여권 얼굴 인증 시스템을 제안한다. 제안하는 지능형 여권 얼굴 인증 시스템은 여권 이미지에서 여권 코드 문자열을 인식하여 여권 사용자의 사진 및 관련 정보를 여권 데이터베이스에서 추출한다. 추출된 출입국자의 사진 및 얼굴과 여권에 부착된 사진 및 얼굴과의 유사도 측정을 통하여 여권 사진의 위조 여부을 판단한다. 이때, 이미지의 유사도 측정을 위해서 다양한 실험을 통한 결과를 종합 분석해 본 결과 사진 영역의 인증에는 Luminance, Edge, RGB 특징이, 얼굴 영역의 인증을 위해서는 Hue, YIQ-I, YCbCr-Cb 특징이 효과적인 것으로 나타났으며 사진 영역의 유사도와 얼굴영역의 유사도가 모두 0.8이상인 경우 정상적인 여권으로 판정하고 그렇지 않은 경우 위조가 되었을 가능성이 있는 여권으로 판정하는 방법을 사용하여 FAR 3.1%, FRR 2.7%의 우수한 결과를 나타내었다.

• Journal of Digital Contents Society
• /
• v.18 no.7
• /
• pp.1323-1331
• /
• 2017

Despite the fact that the copyright has grown into a large-scale business, there are many constant problems especially in image copyright. In this study, we propose an automatic object extraction and classification system for the scanned educational book image by combining document image processing and intelligent information technology like deep learning. First, the proposed technology removes noise component and then performs a visual attention assessment-based region separation. Then we carry out grouping operation based on extracted block areas and categorize each block as a picture or a character area. Finally, the caption area is extracted by searching around the classified picture area. As a result of the performance evaluation, it can be seen an average accuracy of 83% in the extraction of the image and caption area. For only image region detection, up-to 97% of accuracy is verified.

• Proceedings of the IEEK Conference
• /
• 2000.06d
• /
• pp.15-18
• /
• 2000

In this paper, we propose a method for text extraction in the Web images. Our approach is based on contrast detecting and pixel component ratio analysis in mouse position. Extracted data with OCR can be used for real time dictionary call or language translation application in Web browser.

• The Journal of the Korea Contents Association
• /
• v.4 no.3
• /
• pp.125-130
• /
• 2004

This paper proposes a method of the car license plate recognition from digital camera image. Lots of technology advancement has been accomplished for the least several years. The key issue for recognition rate improvement has been the extraction of correct area on the plate. In the previous studies, the information from an edge or an color on a plate hasn't been used but some declination also taken into account in most cases due to the difficulty of area extraction on a tilted plate The proposed method focuses on transforming a slant plate image to the normalized form to be recognized. It shows good robustness on situations defined by a variety of locations, slants and heights of the license plate, because it detects the edge of license plate by using both the color information and linear regression method. The computer simulation shows that the proposed method records 92％ detection rates of license plate and can recognize characters of slant plate with about 50 degrees.