• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.2
• /
• pp.8-18
• /
• 1993

A multi-font, multi-size and high speed character recognition system is designed. The design principles are simpilcity of algorithm, adaptibility, learnability, hierachical data processing and attention by feed back. For the multi-size character recognition, the extracted character images are normalized. A hierachical classifier classifies the feature vectors. Feature is extracted by applying the directional receptive field after the directional dege filter processing. The hierachical classifier is consist of two pre-classifiers and one decision making classifier. The effect of two pre-classifiers is prediction to the final decision making classifier. With the pre-classifiers, the time to compute the distance of the final classifier is reduced. Recognition rate is 95% for the three documents printed in three kinds of fonts, total 1,700 characters. For high speed implemention, a multiprocessor system with the ring structure of four transputers is implemented, and the recognition speed of 30 characters per second is aquired.

• Journal of the Korea Safety Management & Science
• /
• v.17 no.2
• /
• pp.153-160
• /
• 2015

This study tried to investigate the difference of the legibility in different age groups (young and old) with same near visual acuity and to provide preliminary guidelines for the text sizes in the pesticides products based on the legibility experiments. To achieve the objective, legibility evaluation experiments were conducted to test the effects of different age(20s, 50-60s), gender, font type(thick gothic-type and fine gothic-type), thickness (plain and bold), and number of syllables(2 and 3 syllables) in the same near visual acuity(0.6). The results showed that legibility 50s was higher than 20s. And 20s are appealed higher discomfort than 50s. Considering these experimental results, it was recommended that the 14 pt or larger characters (100% readable size) should be used the important information such as toxicity, and the minimum character size was 7 pt (50% readable size) for other information.

• Journal of the Korea Safety Management & Science
• /
• v.12 no.3
• /
• pp.53-60
• /
• 2010

Texts and icons are used to deliver the safety and health information on pesticide labels. This study surveyed the standards or guidelines regarding the presentation formats of texts and icons used in pesticides. Also, the text and icon formats being used in the 50 pesticide products in Korea were investigated in the aspects of size, font (texts only), color. From the guideline survey, it was found that while the guidelines for the text fonts and colors in pesticides were suggested, there was no detailed guidelines about the text size. For the icon presentation, it was regulated that the icon size for the pesticides should be larger than $7mm{\times}7mm$, and this size was equivalent to the size when the reading distance of 28 ㎝ was applied to other two guidelines. From the sample survey of the 50 pesticides, it was found that the small text size being used in the 50 pesticides was 1mm (3pt), and this size was considered to be very difficult to be read, particularly by old people. The minimum size of the texts for the toxicity of the pesticide, the very important safety information, was found to be 1.5 mm, which is considered to be very uncomfortable to be read. The size of icons were varied from 2 to 12 mm, and the total average size was 4.6 mm. The 149 icons from a total 187 icons (80%) were smaller than the icon presentation guideline ($7mm{\times}7mm$), and the reason for the small icon size might be the small area of the label of the pesticides. Thus, more detailed guidelines for the text size and more practical guidelines for the size of icons are required.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.4
• /
• pp.133-146
• /
• 2014

This paper mainly focuses on the design factors of a website targeted at shopping mall websites in China and Korea. We formulated 15 different hypotheses regarding how those design factors are different. The targeted websites are "Gmarket" in Korea and "Taobao" in China. The selected design factors of a shopping mall website are (1) color and design, (2) photo, graphic, and font size, (3) the clarity of information delivery, (4) website structure, (5) screen balance, (6) product display, (7) product harmony, (8) a form of font and icon, (9) visual atmosphere, (10) differentiated color, (11) interesting words, (12) technical skills, (13) entire visual design and atmosphere, (14) navigation, (15) search function for products or information. As a result of analyzing the result using the SAS 9.2 package tool, we figured out that there was a difference between design factors. Additionally, we analyzed this difference and suggested a strategy to design the effective shopping mall web sites.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.1043-1056
• /
• 2017

A styled document is a document that contains diverse decorating functions such as different font, colors, tables and images generally authored in a word processor (e.g., MS-WORD, Open Office). Compared to a plain-text document, a styled document enables a human to easily recognize a logical structure such as section, subsection and contents of a document. However, it is difficult for a computer to recognize the structure if a writer does not explicitly specify a type of an element by using the styling functions of a word processor. It is one of the obstacles to enhance document version management systems because they currently manage the document with a file as a unit, not the document elements as a management unit. This paper proposes a machine learning based approach to analyzing the logical structure of a styled document composing of sections, subsections and contents. We first suggest a feature vector for characterizing document elements from a styled document, composing of eight features such as font size, indentation and period, each of which is a frequently discovered item in a styled document. Then, we trained machine learning classifiers such as Random Forest and Support Vector Machine using the suggested feature vector. The trained classifiers are used to automatically identify logical structure of a styled document. Our experiment obtained 92.78% of precision and 94.02% of recall for analyzing the logical structure of 50 styled documents.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.6
• /
• pp.159-163
• /
• 2008

Recently, typography in Korea has been varied. Among them, after 2005 a new trend caligraphy has appeared and in reality people discussed that this has been occurred as a coincidence. However, as a view of one who is interested in Hangeul typography, born of Hangeul caligraphy is not a coincidence. It is because computerizing of Hangeul has been later than U.S. or Japan. Korea was colonized by Japan and experienced the Korean War and economical independence got late and cultural independence got late as well. Our language, Hangeul was independenced in the beginning of 1990's after computer was introduce and original letter of Hangeul was scanned and basic standard was built. From the end of 80's there were many efforts for independence of Hangeul but there were problems on hardware side than software side. In the beginning of 90's basic fonts such as Myungjo or Gothic of Choi Jung-ho's fonts were set for computer hardware, but later 90's can be called as a time for distributing designed fonts for that new fonts that were planned and applying new ideas on fonts were done. In 2000 the two major font company in Korea Sandoll Communication and Yoon Design Institute made a fonts such as When branding had to be done by designer, typo must be written so adjusting size and moving baseline made old style and caligraphy fonts to be born. These cycling process has been natural motive for nowadays' caligraphy and these two major companies' role has made caligraphy to be popular.

• Journal of The Korean Association of Information Education
• /
• v.6 no.3
• /
• pp.328-337
• /
• 2002

Looking at multimedia education contents from a design point of view, the instructor's design model may differ from the child's understanding model due to gap of the instructor's and child's knowledge. This fact implies it impacts the effectiveness of the education contents. The learning efficiency of Korean typography in WBI for children depends on the font-family, line space, font-size, the age of user, the output device such as the monitor, and other various factors. In this paper, we measured and analyzed on readability of Korean typography in WBI for children by reading speed method. The results of experiments show that readability depends on the font-family of typography, age(grade), and sex of children. In detail, "Goolymche" has the shortest time to be read, and girl and the highest grade students of elementary school have shorter time than boy and the lower grade students. Moreover, we consider the elegance of typography in WBI for holding children's interests because they prefer "Yopseoche". We provide some CSSs in WBI for children based on the experimental results, to used in school fields.

• Proceedings of the IEEK Conference
• /
• 2001.06c
• /
• pp.141-144
• /
• 2001

In this paper, a new algorithm for extracting and recognizing characters from video, without pre-knowledge such as font, color, size of character, is proposed. To improve the recognition rate for videos with complex background at low resolution, continuous frames with identical text region are automatically detected to compose an average frame. Using boundary pixels of a text region as seeds, we apply region filling to remove background from the character Then color clustering is applied to remove remaining backgrounds according to the verification of region filling process. Features such as white run and zero-one transition from the center, are extracted from unknown characters. These feature are compared with a pre-composed character feature set to recognize the characters.

• International Journal of Contents
• /
• v.8 no.4
• /
• pp.95-102
• /
• 2012

In this paper, we present a technique to extract the title areas from book cover images. A typical book cover image may contain text, pictures, diagrams as well as complex and irregular background. In addition, the high variability of character features such as thickness, font, position, background and tilt of the text also makes the text extraction task more complicated. Therefore, we propose a two steps efficient method that uses Histogram of Oriented Gradients and color information to find the title areas. Firstly, text localization is carried out to find the title candidates. Finally, refinement process is performed to find the sufficient components of title areas. To obtain the best result, we also use other constraints about the size, ratio between the length and width of the title. We achieve encouraging results of extracted title regions from book cover images which prove the advantages and efficiency of the proposed method.

• Annual Conference on Human and Language Technology
• /
• 1990.11a
• /
• pp.183-190
• /
• 1990

본 논문에서는 인쇄체 한글문자를 실용적으로 인식하기 위하여 고안된 계층적 신경망을 소개하고, 이를 다중활자체의 한글문자를 인식하는 문제에 적용하였다. 이 신경망은 입력된 문자영상을 6가지의 유형으로 분류한 후, 해당 유형을 처리하는 신경망에서 실제 문자를 인식하도록 구성되었다. 또한 각 신경망을 모든 입력영상의 모든 출력노드에 대해 고르게 학습시키기 위하여 Backpropagation 알고리즘을 개선한 Descending Epsilon 알고리즘을 도입하였다. 그 결과 사용빈도수가 높은 한글 520자에 대해 94.4 - 98.4%의 인식률을 얻음으로써 본 논문에서 제안한 시스템이 다양한 활자체로 이루어진 실제 문서인식시스템의 문자인식부에 효과적으로 사용될 수 있음을 제시하였다.