• Korean Journal of Cognitive Science
• /
• v.1 no.1
• /
• pp.27-76
• /
• 1989

• The Journal of the Korea Contents Association
• /
• v.7 no.10
• /
• pp.146-156
• /
• 2007

This study is possible because, first, "rammatology" and typography are connected by language. Second, I believe that language in typography should be comprehended in a relationship between peripheral studies such as history, philosophy, literature, and linguistics. Finally, "grammatology" is an excellent subject for experiments which attempt to rehabilitate the right of language, which is dependent on sound within the metaphysical boundary. This study is formed on the idea that the study of typography should be based on language, which is the basic foundation, and that this is possible in investigations related to peripheral studies. The main purpose of typography is an accurate delivery of meaning. However, the dynamics of creative thoughts towards advancement mainly depends on experiment, thus we cannot limit the purpose of typography as a means of communication to exchange meanings. For these reasons, I admit that interest in peripheral studies -which are not yet approached in the study on typography -take precedence.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.73-75
• /
• 2022

Scene Text Recognition is a technology used in the field of artificial intelligence that requires manless robot, automatic vehicles and human-computer interaction. Though scene text images are distorted by noise interference, such as illumination, low resolution and blurring. Unlike previous studies that recognized only English, this paper shows a strong recognition accuracy including various characters, English, Korean, special character and numbers. Instead of selecting only one class having the highest probability value, a candidate word can be generated by considering the probability value of the second rank as well, thus a method can be corrected an existing language misrecognition problem.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.378-380
• /
• 1998

최근의 HPC같은 초소형 컴퓨터의 발달은 더 자연스럽고 더 사용하기 편한 입출력 시스템을 요구하게 되었다. 본 논문에서는 HPC상에서의 흘림한글을 인식할 수 있는 인식 시스템을 구현한 것을 주제로 하였다. 본 시스템은 획을 인식의 기본 단위로 취급하며, 획 인식을 위하여 ART-1신경망을 사용하였으며, 글자인식을 위해 HMM의 각 스테이트를 탐색하는 방법을 사용하였다. 본 논문에서는 이 시스템을 HPC상에서 구현하였고 좋은 실험결과를 얻었다.

• The Science & Technology
• /
• v.35 no.10 s.401
• /
• pp.22-23
• /
• 2002

국내 최초로 한국어를 기반으로 한 음성인식종합기술을 개발해 기술의 상품화에 성공한 (주)SL2는 KAIST 학생벤처기업 1호이다. SL2는 사람의 목소리를 컴퓨터로 분석한 후 이를 글자로 출력하는 음성인식을 10만단어까지 가능케하는 기술을 보유해 업계 최고 수준을 자랑하고 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.460-462
• /
• 2001

컴퓨터 이미지처리는 여러 분야에서 응용되고 있는데 어떤 특성을 만족하는 객체들의 계수를 자동으로 분류시키는 생물학분야, 편지봉투나 일반양식에 인쇄되어 있는 글자를 자동으로 검출하고 인식하며 초음파검사 혹은 X-Ray 촬영에서 이미지를 획득하여 향상시키는 의료분야, 지문 및 얼굴인식 등에 이용되고 있다. 최근 몇 년 동안 이미지인식, 형태론, 이미지데이터 압축에 관한 연구가 진전되면서 본 연구에서 형태론적인 기법을 사용하여 문자인식을 위한 전처리 혹은 후처리 단계에서 사용되는 이미지향상을 위해서 팽창, 침식, 골격화의 3단계를 적용하고 기존의 연구 방법과 비교하여 이미지획득 시간을 줄이고 이미지를 향상시켰다.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.233-236
• /
• 2000

최근 수많은 인쇄된 문서들이 HTML과 같은 디지털 문서로 바뀌고 있으며 이를 자동으로 변환해 주는 문자인식 기술에 대한 관심이 증가하고 있다. 본 논문에서는 그림과 글자가 공존하는 문서에서 자동으로 문자영역을 추출해서 문자를 인식하는 방법을 제안한다. 우선 입력문서는 유사한 칼라로 이루어진 영역들로 나누어진 뒤 휴리스틱 룰에 의해 문자후보 영역과 비 문자 영역으로 나누어진다. 그 다음 이들 문자후보영역들은 문자인식기를 이용하여 문자 혹은 문자의 일부분으로 인식된다. 제안된 방법으로 여러 문서들에 대하여 실험한 결과를 보이며 그 성능을 평가한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.248-251
• /
• 2016

This thesis has discussed how to recognize and convert raised letters on tire to data and collect such data. Unlike the existing recognition system, the system presented by this thesis recognizes raised letters on tire through detecting letters after converting the Kinect camera image into image data in the preprocessing stage. After then, numbers and letters are analyzed through image improvement by use of binary images, noise filter, etc. In the recognition stage, letter distinction is used and raised letters on tire are recognized 100% through correction of errors by way of the correction algorithm for tire data recognition errors. In this paper it will be the development of a method of recognizing characters and the tire technology. Although there are many ways to the already recognized characters, Tire characters requires a technique different from the more general character recognition. For this reason and to develop additional technical methods and algorithms for character recognition.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.5
• /
• pp.128-137
• /
• 2009

This research aimed at providing the direction for future improvement of broadcast subtitle design by identifying the user perception on broadcast subtitles of Terrestrial DMB. First, we contemplated on the existing research about broadcast subtitles. As a result, we found out that broadcast subtitle plays an essential role when watching a program and that the minimization of letters, clear fonts and color arrangement, rearrangement of components were identified as main factors that increase readability. However, the terrestrial DMB user evaluation on image quality, composition, color and brightness was not that positive. In addition, the evaluation on the readability of broadcast subtitle, information transmission, subtitle composition, letter font, number and size of letter was more negative. Developing a subtitle design which considers user convenience and suits terrestrial DMB screens in aspects of broadcast subtitle form and composition, limit of information, clearness and size of the letter is required.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.17-25
• /
• 2023

In this paper, we propose a lightweight deep learning architecture tailored for efficient text detection in fashion design sketch images. Given the increasing prominence of Digital Transformation in the fashion industry, there is a growing emphasis on harnessing digital tools for creating fashion design sketches. As digitization becomes more pervasive in the fashion design process, the initial stages of text detection and recognition take on pivotal roles. In this study, a lightweight network was designed by building upon existing text detection deep learning models, taking into consideration the unique characteristics of apparel design drawings. Additionally, a separately collected dataset of apparel design drawings was added to train the deep learning model. Experimental results underscore the superior performance of our proposed deep learning model, outperforming existing text detection models by approximately 20% when applied to fashion design sketch images. As a result, this paper is expected to contribute to the Digital Transformation in the field of clothing design by means of research on optimizing deep learning models and detecting specialized text information.