• 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 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.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.6
• /
• pp.405-413
• /
• 2009

Tagging system is the system that allows internet users to assign new meta-data which is called tag to article, photo, video and etc. for facilitating searching and browsing of web contents. Tag cloud, a visual interface is widely used for browsing tag space. Tag cloud selects the tags with the highest frequency and presents them alphabetically with font size reflecting their popularity. However the conventional tag selection method includes known weaknesses. So, we propose a novel tag selection method Freshness, which helps to find fresh web contents. Freshness is the mean value of Kullback-Leibler divergences between each consecutive change of tag co-occurrence probability distribution. We collected tag data from three web sites, Allblog, Eolin and Technorati and constructed the system, 'Fresh Tag Cloud' which collects tag data and creates our tag cloud. Comparing the experimental results between Fresh Tag Cloud and the conventional one with data from Allblog, our one shows 87.5% less overlapping average, which means Fresh Tag Cloud outperforms the conventional tag cloud.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.3
• /
• pp.47-65
• /
• 2014

VMS(Variable Message Sign) helps drivers to choose their path to destinations on roadways. Some types of VMS often provide traffic information with clearly visible and comprehensible graphical route diagrams. Currently many diagramed types of VMS are installed on urban arterial and highways. This type of VMS surely enhances drivers' ability to comprehend traffic route information while they are driving on the roadway. Nevertheless, some of them are presented with so much information and design elements and they sometimes lead to decline of drivers' comprehensible level for traffic information. Drivers would fail to decide their preferable route in this state of information overflow. The purpose of this paper is to develop a drivers preference model for effective design principle including size and height of displaying font, and the amount of information in the route diagram considering driving speed, sex and age of the driver. This model is developed using structural equation modeling techniques. This model considers driver's emotional factor and, human factor and design component of route diagram. To collect data, we built driving simulator which is able to replicate real driving condition. 72 people who participated in the simulation were selected considering gender and age. The developed model showed that the amount of information, and visibility are more influential factors to the drivers' preference of the route diagram on VMS than design elements such as the shape and the font of the diagram.

• Journal of the Korean Society of Food Culture
• /
• v.30 no.2
• /
• pp.190-196
• /
• 2015

Restaurants implement menu labeling to provide nutritional information to customers in an attempt to help customers select healthy menu items. Considering the increase in food-away-from-home consumption, the purpose of this study was to identify motivators and barriers in restaurant customers regarding use of menu labeling. Data were collected from a survey on restaurant customers in Seoul, Korea. The findings of this study indicate that customers used menu labeling for health reasons. However, barriers to using menu labeling were identified as small font size, difficulty in locating nutritional information display, and difficulty in interpreting nutritional information. In addition, they also suggested expanding the scope of menu labeling for restaurants by including chain restaurants with less than 100 units. The findings of this study offer strategies for the government to improve menu labeling practices for customers.

• Journal of the Korean Society for Library and Information Science
• /
• v.43 no.2
• /
• pp.171-192
• /
• 2009

To date, large print books have been published in Korea only for partially-sighted persons. However, as a recent explanation in the Reading Culture Promotion Act indicates, people who cannot use reading materials due to old age are now also considered to be reading-handicapped. This study is aimed at investigating the proper characteristics for the improvement of print publication for elderly people using data from a survey of 80 elderly persons. This study identified elderly people's preferences on the major characteristics of large format books such as font style, character size, line width, and page layout, derived from both quantitative and quantitative research methods.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.26 no.11
• /
• pp.1627-1637
• /
• 2002

The objective of this study is to analyze the images of basic bodice apparel displayed on the PC monitor connected to digital camera, and to suggest a more efficient alternative method that enables the expert judge to evaluate the subject both traditional method and displayed images on the computer. Appearance evaluation has traditionally been assessed using subjective method dependent upon expertjudges' senses with the naked eye after direct observation of a living model. This research faces the new challenge focusing on the sensory evaluation of appearance for basic apparel patterns, which can overcome the space and time limitations of the traditional feet methods. A total fifteen basic bodice garment (3 types of $pattern\;{\times}\;5$ subject) were constructed with same fabrics. The appearance evaluation items consist of lg questions of upper torso. The image takes font, back and side view of the dressed subject with three different situations. Data was analyzed using percentiles, standard deviation T-test and ANOVA. Taken together, the present result of appearance evaluation through digital camera image shows that there is a significant difference ($p{\leq}.001$) in the response to the placement of the neckline, the waist & shoulder dart, the general ease of the bust & waist area, the side seam, the perpendicular of the waist hem and general acceptability between the above three different situations; the image in the condition of greenish yellow background with front light showed the highest score through all questions. These results depend on the kind of background colors with the light.

• 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.

• International journal of advanced smart convergence
• /
• v.4 no.2
• /
• pp.94-102
• /
• 2015

Optical camera communication (OCC) is an extension of Visible Light Communication. Different from traditional visible light communication, optical camera communications is an almost no additional cost technology by taking the advantage of build-in camera in devices. It was became a candidate for communication protocol for IoT. Camera module can be easy attached to IoT device, because it is small and flexible. Furthermore almost smartphone equip one or two camera for both back and font side with high quality and resolution. It can be utilized for receiving the data from LED or positioning. Actually, OCC combines illumination and communication. It can supply communication for special areas or environment where do not allow Radio frequency such as hospital, airplane etc. There are many concept and experiment be proposed. In this paper we proposed utilizing Android smart-phone camera for receiver and introduce new approach in modulation scheme for LED at transmitter. It also show how Manchester coding can be used encode bits while at the same time being successfully decoded by Android smart-phone camera. We introduce new data frame format for easy decoded and can be achieve high bit rate. This format can be easy to adapt to performance limit of Android operator or embedded system.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.6_2
• /
• pp.699-706
• /
• 2021

Existing license plate recognition system is used as an optical character recognition method, but a method of using deep learning has been proposed in recent studies because it has problems with image quality and Korean misrecognition. This requires a lot of data collection, but the collection of license plates is not easy to collect due to the problem of the Personal Information Protection Act, and labeling work to designate the location of individual license plates is required, but it also requires a lot of time. Therefore, in this paper, to solve this problem, five types of license plates were created using a virtual Korean license plate generation program according to the notice of the Ministry of Land, Infrastructure and Transport. And the generated license plate is synthesized in the license plate part of collectable vehicle images to construct 10,147 learning data to be used in deep learning. The learning data classifies license plates, Korean, and numbers into individual classes and learn using YOLOv5. Since the proposed method recognizes letters and numbers individually, if the font does not change, it can be recognized even if the license plate standard changes or the number of characters increases. As a result of the experiment, an accuracy of 96.82% was obtained, and it can be applied not only to the learned license plate but also to new types of license plates such as new license plates and eco-friendly license plates.