• 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.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06b
• /
• pp.275-280
• /
• 2008

본 논문에서는 한글 글립(glyph)의 형태학적 분석을 통해 자동으로 획을 분할하는 방법을 제안한다. 제안된 방법은 thinning된 한글 글립의 골격(skeleton) 이미지를 기반으로, 획 분리, 획 병합, 그리고 획 볼륨 복원의 세가지 단계를 거쳐 한글의 기본 획들을 추출해 낸다. 실험 결과, 트루타입폰트(TrueType Font)에 대해서는 80%, 손글씨(Handwriting) 글립에 대해서는 72%의 획 분할 정확도를 보였다. 본 논문에서 제안한 방법으로 획득된 획 정보를 이용하여, 향후 한글 손글씨 생성을 위한 연구를 하고자 한다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.12
• /
• pp.894-906
• /
• 2006

There are many difficulties to develop a mobile contents due to many constraints on mobile environments. It is difficult to make a good mobile contents with only visual reduction of existing contents on wire Internet. Therefore, it is essential to devise the data representation and to develop the authoring tool to meet the needs of the mobile contents market. We suggest the compact mobile contents to learn Chinese characters and developed its authoring tool. The animation which our system produces is realistic as if someone writes letters with pen or brush. Moreover, our authoring tool makes a user generate a Chinese character animation easily and rapidly although she or he has not many knowledge in computer graphics, mobile programming or Chinese characters. The method to generate the stroke animation is following: We take basic character shape information represented with several contours from TTF(TrueType Font) and get the information for the stroke segmentation and stroke ordering from simple user input. And then, we decompose whole character shape into some strokes by using polygonal approximation technique. Next, the stroke animation for each stroke is automatically generated by the scan line algorithm ordered by the stroke direction. Finally, the ordered scan lines are compressed into some integers by reducing coordinate redundancy As a result, the stroke animation of our system is even smaller than GIF animation. Our method can be extended to rendering and animation of Hangul or general 2D shape based on vector graphics. We have the plan to find the method to automate the stroke segmentation and ordering without user input.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.470-474
• /
• 2019

한글 OCR 성능을 높이기 위해 딥러닝 모델을 활용하여 문자인식 부분을 개선하고자 하였다. 본 논문에서는 폰트와 사전데이터를 사용해 딥러닝 모델 학습을 위한 한글 문장 이미지 데이터를 직접 생성해보고 이를 활용해서 한글 문장의 OCR 성능을 높일 다양한 모델 조합들에 대한 실험을 진행했다. 딥러닝 모델은 STR(Scene Text Recognition) 구조를 사용해 변환, 추출, 시퀀스, 예측 모듈 각 24가지 모델 조합을 구성했다. 딥러닝 모델을 활용한 OCR 실험 결과 한글 문장에 적합한 모델조합은 변환 모듈을 사용하고 시퀀스와 예측 모듈에는 BiLSTM과 어텐션을 사용한 모델조합이 다른 모델 조합에 비해 높은 성능을 보였다. 해당 논문에서는 이전 한글 OCR 연구와 비교해 적용 범위를 글자 단위에서 문장 단위로 확장하였고 실제 문서 이미지에서 자주 발견되는 유형의 데이터를 사용해 애플리케이션 적용 가능성을 높이고자 한 부분에 의의가 있다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.469-471
• /
• 2020

방송의 디지털화에 따른 비장애인 대비 소외 계층의 정보 접근성의 부족은 소외 계층에 대한 정보 격차를 심화시킬 수 있다. 이에 캐릭터 수화 방송, 자막 방송 등 장애인을 위한 방송 서비스의 양적, 질적 개선에 관한 연구가 진행되고 있다. 자막 방송 서비스의 경우, 관련 법령에 따라 서비스를 실시하고 있으며 지상파 UHD 방송의 경우에도 본 방송을 시작한 이래 폐쇄 자막 서비스 시스템을 구축하여 서비스를 제공하고 있다. 이러한 기존 자막 서비스는 텍스트 형태의 단조로운 내용 전달 방식이어서 다양한 스타일로 풍부하게 내용을 전달하는 것에 대한 요구가 있다. 이에 본 논문에서는 지상파 UHD 방송을 대상으로 개선된 형태의 자막 서비스인 감성자막 서비스를 소개하고 이를 위한 감성 자막 편집기 기술 개발에 대하여 다룬다. 감성자막 서비스는 화자의 감정 정보를 자막 메타데이터에 추가적으로 제공하여, 감정에 따라 다양한 이모티콘이나 다른 종류의 폰트 스타일로 자막 서비스가 가능하게 하는 서비스이다. 감성자막 편집기는 이러한 감성 자막 메타데이터를 추가, 편집하고 감성자막 파일로 생성하기 위한 시스템으로, 지상파 UHD 송출 시스템 및 폐쇄 자막 표준을 고려하여 개발하였다.

• 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.

• 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.

• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.39 no.2
• /
• pp.66-75
• /
• 2002

This paper presents an adaptive back propagation algorithm that its able to enhancement for the learning efficiency with updating the learning parameter and varies the number of hidden layer node by the generated error, adaptively. This algorithm is expected to escaping from the local minimum and make the best environment for the convergence of the back propagation neural network. On the simulation tested this algorithm on three learning pattern. One was exclusive-OR learning and the another was 3-parity problem and 7${\times}$5 dot alphabetic font learning. In result that the probability of becoming trapped in local minimum was reduce. Furthermore, the neural network enhanced to learning efficient about 17.6%~64.7% for the existed back propagation. 

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.9
• /
• pp.1070-1075
• /
• 2019

In this paper, I propose an algorithm to generate an input digit image for a digit recognition system by detecting a digit string in an image and segmenting the digits constituting the digit string. The proposed algorithm detects blobbed digit string through blob detection, designates a digit string area and corrects digit string skew using the detected blob information. And the proposed algorithm corrects the digit skew and determines the boundary points for the digit segmentation in the corrected digit sequence using three CPgraphs newly defined in this paper. In digit segmentation experiments using the image group including digit strings printed with a range of the font sizes and the image group including handwritten digit strings, the proposed algorithm successfully segments 100% and 90% of the digits in each image group.

• Journal of The Korean Association of Information Education
• /
• v.6 no.2
• /
• pp.212-224
• /
• 2002

The purpose of this study is to design and implement the web-based counselling system in aspects of learning type and method by PBL in Constructivism. Learning problem is classified into learning difficulty, learning disability, emotional disability, memory/cognition, ADHD(attention deficit hyperactivity disorder), classwork, and school circumstance. Counselling for individual, offline visit, parents, electronic mail, and group with professionals who were involved in the design and implementation of learning counselling were used to supplement the clinic process for learning. The result of this study aims to administer the distance counselling system for learning without prominent computer knowledge. Each learner can solve diverse private learning problems which can be widely applied for situated learning. Also these online strategies by PBL indicate that providing learning types and methods are critical for overall understanding in online counselling system for learning.