• The KIPS Transactions:PartA
• /
• v.15A no.2
• /
• pp.75-86
• /
• 2008

Black/white comic rendering is one of the researches in the field of non-photorealistic rendering(NPR). Black/white comics have been produced manually as yet. But these previous systems require lots of time and manual work. So we propose the COmics Rendering system on VIdeo Stream (CORVIS) which transforms video streams into black/white comic cuts. Stylized font, one of comic representations, can be used to express onomatopoeic words and mimetic dialogue exaggeratively. But current comic generation systems do not provide enough effects of stylized font. This paper proposes a model for stylized fonts to express various effects. Effects of stylized fonts we proposed include geometric deformations. Thus we could represent stylized fonts on the still cut of movies and the background texture on a cuts of plain black/white comics. The final quality of our system produced is good enough to compare with manual black/white comics.

• KIISE Transactions on Computing Practices
• /
• v.21 no.11
• /
• pp.699-705
• /
• 2015

Thousands of glyphs are necessary for Hangul font generation. It is mandatory to generate the required glyphs before producing Hangul font. This paper, entitled "Multiple Master Glyph Algorithm", presents an process that generates a target number of glyphs automatically from a very small number of glyphs by using a combination rule setting and a glyph interpolation method. A font editor, which is able to generate Hangul glyphs or fonts, is developed based on this algorithm. The editor generates a target number of fundamental glyphs automatically by using a combination rule setting and four master glyphs, which can be set up by a user. The automatically generated glyphs can be used to generate a target font by combining KSX1001 standard Hangul 2350 characters or Unicode standard Hangul 11172 characters automatically. The efficiency of the proposed Hangul editor is analyzed quantitatively in this paper through application to several commercial typefaces.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.255-256
• /
• 2019

한글 폰트를 만드는 데는 자음+모음 조합으로 약 11,500자 정도의 글자가 필요하다. 디자이너가 글자 하나씩 전부 디자인 하는 것도 굉장한 부담요소이고, 한글폰트를 제작하는데 있어 3개월 이상의 소요 기간과 3000만 원 이상의 비용부담 또한 무시 못 할 요소이다. 게다가 카피라이트 폰트에 대한 저작권 문제 또한 골칫거리다. 그래서 이를 최소한으로 하고자 딥 러닝의 방식중 하나인 GAN(생성적 적대 신경망)을 통해서 디자이너가 399자만 작성하고 나머지는 컴퓨터가 디자이너의 폰트 디자인을 인식하고 자동으로 만들어 주는 프로그램을 고안하였다.

• Proceedings of the KIEE Conference
• /
• 1989.07a
• /
• pp.461-464
• /
• 1989

This paper deals with the computer generation of Korean characters by the structural coding which results in higher flexibility and compactness. Our method by which Korean characters are designed is characterized as follows : The list of primitives for Korean text is extracted by structural coding rule, and the knowledge-base is used for handling various primitives.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.27 no.5
• /
• pp.802-807
• /
• 1990

In this thesis, we represent the study on the design and implementation of the hangul graphic board which generate bit map font data from the boundary information of korean or chines fonts. The implemented graphic board consists of a TMS34010 Graphic System Processor (GSP) and two TMS320C25 Digital Signal Processor (DSP), and there is shared memory which consists of two memory blocks with same address for which is possible parallel processing between two processors. And in using DSP, we propose an efficient algorithm for calculation of Bezier curve which require much times to calculate bit map data font from the boundary information.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06b
• /
• pp.275-278
• /
• 2010

이 논문에서는, 기존에 존재하는 글자체의 커브 데이터를 분석하여 같은 글자를 붓글씨로 서예를 하듯이 다시 써낸 듯한 효과를 낼 수 있는 방법을 제안한다. 글자를 형성하는 위상적인 뼈대를 커브로 쪼개어, 글자 하나를 여러 획으로 분리하여 표현한 후에, 각 획에 해당하는 커브의 차원 수와, 길이와 곡률을 이용하여 붓의 궤적을 자동적으로 생성해 내는 방법이다. 붓의 궤적이 표현될 방법을 기존 글자 데이터를 이용해서 어떻게 조작 경로를 자동적으로 만들어 붓글씨 팔적을 생성해낼 것인지가 풀어내어야 할 문제이다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.12
• /
• pp.499-508
• /
• 2022

A lot of research has been carried out on the Hangeul generation model using deep learning, and recently, research is being carried out how to minimize the number of characters input to generate one set of Hangul (Few-Shot Learning). In this paper, we propose a CKFont2 model using only 14 letters by analyzing and improving the CKFont (hereafter CKFont1) model using 28 letters. The CKFont2 model improves the performance of the CKFont1 model as a model that generates all Hangul using only 14 characters including 24 components (14 consonants and 10 vowels), where the CKFont1 model generates all Hangul by extracting 51 Hangul components from 28 characters. It uses the minimum number of characters for currently known models. From the basic consonants/vowels of Hangul, 27 components such as 5 double consonants, 11/11 compound consonants/vowels respectively are learned by deep learning and generated, and the generated 27 components are combined with 24 basic consonants/vowels. All Hangul characters are automatically generated from the combined 51 components. The superiority of the performance was verified by comparative analysis with results of the zi2zi, CKFont1, and MX-Font model. It is an efficient and effective model that has a simple structure and saves time and resources, and can be extended to Chinese, Thai, and Japanese.

• Proceedings of the Korean Society of Tribologists and Lubrication Engineers Conference
• /
• 2002.10b
• /
• pp.115-116
• /
• 2002

This study was carried out as one of efforts to overcome difficulties in air bearing design due to low stiffness and low damping. Hydrodynamic effects on hydrodynamic stiffness of a fluid film in a high speed air bearing with tow-row air sources are investigated. The hydrodynamic effects by the high speed over DN 1,000,000 and eccentricity of a proceeding which are not considered in conventional design of an air bearing need to be reconsidered. The hydrodynamic effects, which dominantly influence on the load capacity of air bearing, are caused mainly by proceeding speed, eccentricity, and the source positions. The two-row source arrangement in the air bearing produces quite unique hydrodynamic effects with respect to pressure distribution of the air film. Optimal arrangement of the two-row sources improves performance of an air bearing in film reaction force and loading capacity of high speed spindles. This study compares the pressure distribution by numerical simulation as a function of eccentricity of proceeding and the source positions. The air source position 1/7L form one end of an air bearing was found to be superior to source position of 1/4L. The dynamic stiffness were obtained using a two-dimensional cutting method which can directly measure the cutting reaction forces and the displacements of the spindle in two directions using a tool dynamometer and transducer sensors. Heat generation in the air film can not be negligible over the speed of DN 2,000,000. In order to analysis effects of heat generation on the characteristics of air bearing, high cooling bearing spindle and low cooling bearing spindle were tested and compared. Characteristics of the frequency response of shaft and motion of run out errors were different for the spindle. The test results show that, in the case of low cooling bearing spindle, the stiffness became smaller due to heat generation. The results, which were obtained for high speed region, may be used as a design information for spindle which can be applied to precision devices such as ultra precision grinding and ultra high speed milling.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.275-278
• /
• 2020

In this paper, we proposed a deep learning based character recognition system through virtual data generation. In order to secure the learning data that takes the largest weight in supervised learning, virtual data was created. Also, after creating virtual data, data generalization was performed to cope with various data by using augmentation parameter. Finally, the learning data composition generated data by assigning various values to augmentation parameter and font parameter. Test data for measuring the character recognition performance was constructed by cropping the text area from the actual image data. The test data was augmented considering the image distortion that may occur in real environment. Deep learning algorithm uses YOLO v3 which performs detection in real time. Inference result outputs the final detection result through post-processing.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1184-1191
• /
• 2021

In a pop song, the creator's intention is communicated to the user through music and lyrics. Lyric meaning is as important as music, but in most cases lyrics are delivered to users in a static form without non-verbal cues. Providing lyrics in a static text format is inefficient in conveying the emotions of a music. Recently, lyrics video with kinetic typography are increasingly provided, but producing them requires expertise and a lot of time. Therefore, in this system, the emotions of the lyrics are found through the analysis of the text of the lyrics, and the deep learning model is trained with the data obtained by converting the melody into a Mel-spectrogram format to find the appropriate emotions for the music. It sets properties such as motion, font, and color using the emotions found in the music, and automatically creates a kinetic typography video. In this study, we tried to enhance the effect of conveying the meaning of music through this system.