• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.15-20
• /
• 2006

This paper describes research on intelligent game character through performance enhancements of physics engine in computer games. The algorithm that recognizes the physics situation uses momentum back-propagation neural networks. Also, we present an experiment and its results, integration methods that display optimum performance based on the physics situation. In this experiment on integration methods, the Euler method was shown to produce the best results in terms of fps in a simulation environment with collision detection. Simulation with collision detection was shown similar fps for all three methods and the Runge-kutta method was shown the greatest accuracy. In the experiment on physics situation recognition, a physics situation recognition algorithm where the number of input layers (number of physical parameters) and output layers (destruction value for the master car) is fixed has shown the best performance when the number of hidden layers is 3 and the learning count number is 30,000. Since we tested with rigid bodies only, we are currently studying efficient physics situation recognition for soft body objects.

• Journal of the Korean Society of Costume
• /
• v.57 no.3 s.112
• /
• pp.1-13
• /
• 2007

The Purpose of this study is to analyze the kinds and positions of tattoo patterns on the body in tattoo-related internet communities and professional web sites. for this purpose, 1,892 tattoo patterns were analyzed by sex(man and woman). The results were as fellows; First, animal patterns(30.2%) occupied most, followed by character patterns(24.1%), geometric patterns(13.0%), natural patterns(10.3%), plant patterns(4.7%), mixed patterns(2.5%), and artificial patterns(2.2%). In patterns, dragon(10.3%) occupied most, followed by star(8.7%), trival(8.6%), woman(7.6%), skeleton(4.9%), and letter(4.8%). Second, men's preference to pattern groups included animal patterns(30.8%), character patterns (28.3%), geometric patterns (14.6%), and natural patterns(6.0%). Among patterns, dragon(13.4%) was the most frequent, followed by trival(10.9%), woman(10.7%), and skeleton(7.1%). Women's preference to patterns groups included animal patterns(31.4%), natural patterns(17.3%), character patterns(17.2%), geometric patterns(10.5%), and plant patterns(10.0%). Among patterns, star(15.3%) was the most frequent, followed by butter- fly(10.5%), elf(9.2%), and dragon(9.2%). Third, the positions of tattoos on the body included upper arm(26.6%), shoulder(10.8%), back(10.5%), the wrist(10.0%), the calf(7.5%), back bottom(7.0%) and the breast(6.3%). While men's preference to pattern positions included upper arm(38.2%), the wrist(13.7%), back(10.5%), the calf(9.4%), and shoulder(8.0%), women's preference to positions included back bottom(17.7%), shoulder(15.5%), back(10.5%), front bottom(8.2%), and the breast(7.8%).

• Korean Journal of Environmental Biology
• /
• v.35 no.4
• /
• pp.533-544
• /
• 2017

In this study, the fauna of freshwater fish were investigated from March until October of 2016 in 24 independent streams in the Jeonnam and Jeonbuk provinces, which were flowing into the Yellow Sea. The health of the aquatic ecosystem in those streams was assessed through the biological and abiological character index (BAc index). During the surveyed period, a total of 4,127 individuals were collected; they belonged to 59 species of 44 genera in 18 families. The most dominant species identified was Zacco platypus, and 12 species were endemic species of Korea, including Rhodeus uyekii. The BAc index indicated a statistically significant correlation (p<0.01 or 0.05). The stage distribution of the aquatic ecosystem health assessment showed the highest rating of 41.7% at the "Fair" stage. The "Good" and "Poor" stages accounted for 20.8% while the "Excellent" stage took up 16.7%. It has been confirmed that the aquatic ecosystem health of independent streams is influenced more by the abiological character index as a consequence of geographical characteristics and artificial/natural limiting factors, than by the biological character index.

• Journal of Convergence for Information Technology
• /
• v.12 no.1
• /
• pp.163-170
• /
• 2022

With the advent of the 4th Industrial Revolution, solutions that apply AI technology are being actively developed. Since 2017, the introduction of business automation solutions using AI-based Robotic Process Automation (RPA) has begun in the financial sector and insurance companies, and recently, it is entering a time when it spreads past the stage of introducing RPA solutions. Among the business automation using these RPA solutions, it is very important how accurately textual information in the document is recognized for business automation using various documents. Such character recognition has recently increased its accuracy by introducing deep learning technology, but there is still no recognition model with perfect recognition accuracy. Therefore, in this paper, we checked how much accuracy is improved when pre- and post-processor technologies are applied to deep learning-based character recognition engines, and implemented RPA recognition engines and linkage technologies.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.33 no.4
• /
• pp.87-98
• /
• 2015

The natural monument forests (no.374) located at Pyeongdae-ri in Jeju island are described and classified by using phytosociological methods and numerical analysis. The purpose of this paper is to identify the ecological character of Torreya nucifera forests between natural habitat and artificial habitat, as well as their spatial and phytogeographical distribution in the Korea. The comparison of forests between Pyeongdae-ri and other regions was analyzed by using a non-metric multidimensional scaling analysis (NMDS) and hierarchical clustering. On the basis of the 12 phytosociological $relev{\acute{e}}s$, the vegetation of T. nucifera dominant forest in Jeju island was arranged in one syntaxon (Alangium platanifolium-Torreya nucifera community included typicum and one subcommunity) within Camellietea. The community of T. nucifera dominant forests were characterized floristically and ecologically. We discussed diagnostic species with references, and proposed a few important diagnostic species (Ilex crenata for. microphylla, Acer palmatum, Zingiber mioga, Mercurialis leiocarpa, Osmorhiza aristata, Mecodium wrightii etc.) to explain condition of the habitat and synecological character. The communities were described by concerning their edaphical and syndynamical niche; we discussed their total distribution in Korea. In most forests they are widespread in Korean peninsular and their distribution is primarily determined by artificial plantation and periodical management. The forests consisted of T. nucifera have developed from natural environment element and artificial management. As a result they have very unique characters with the floristic, structural characterization and distribution. Furthermore, we identified that they need to apposite management for sustainability.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.1
• /
• pp.9-17
• /
• 2021

In this paper, we present a simple and fast supervised learning framework based on model predictive control so as to learn motion controllers for a physic-based character to track given example motions. The proposed framework is composed of two components: training data generation and offline learning. Given an example motion, the former component stochastically controls the character motion with an optimal controller while repeatedly updating the controller for tracking the example motion through model predictive control over a time window from the current state of the character to a near future state. The repeated update of the optimal controller and the stochastic control make it possible to effectively explore various states that the character may have while mimicking the example motion and collect useful training data for supervised learning. Once all the training data is generated, the latter component normalizes the data to remove the disparity for magnitude and units inherent in the data and trains an artificial neural network with a simple architecture for a controller. The experimental results for walking and running motions demonstrate how effectively and fast the proposed framework produces physics-based motion controllers.

• Journal of Korean Elementary Science Education
• /
• v.36 no.3
• /
• pp.286-301
• /
• 2017

Nowadays, the public as well as science educators pays much attention to the fourth industrial revolution and wonders what will happen to the societies in the future. Thus, this study aimed at predicting the education environment which will be brought from the fourth industrial revolution, and suggesting the solutions or tasks to be investigated in science education. Through the literature review, this study categorized the major changes of future society into a wild fluctuation of job market, the shift from possession-based economy to sharing economy, post-urbanized and distributed system, and the crisis of dehumanization. According to the four major changes, this study predicted the future environment that will occur to the educational system. First, the students should the competences necessary for the future and the school curriculum will be changed in terms of width and depth. Second, sharing economy may bring about the open platform similar to MOOC (Massive Open Online Course) or TED. Third, the manifestation of artificial intelligence in education will enable the individual and paced learning, and thanks to the change, the concept of distributed cognition will be more focused in education research. Fourth, the collaborative learning and character education should be more stressed to resist the dehumanization. This study suggests relevant tasks and issues that should be tackled for the successful change in primary and secondary schools.

• Journal of Fisheries and Marine Sciences Education
• /
• v.28 no.3
• /
• pp.653-660
• /
• 2016

Snakehead, Channa argus were intraperitoneally infected with Nocardia seriolae. at the concentrations of $1.5{\times}10^7cfu/m{\ell}$ or $1.5{\times}10^8cfu/m{\ell}$. The infected fish were kept in aquaria at $25^{\circ}C$ and $30^{\circ}C$ for 3 weeks. Clinical signs and mortality were monitored daily to evaluate the virulence. All artificially infected fish showed the same clinical sign found in naturally infected fish. All the fish infected with $1.5{\times}10^8cfu/m{\ell}$ of N. seriolae died within 24days. N. seriolae showed higher virulence to snakehead at the temperature $30^{\circ}C$. Internal lesions such as whitish nodules in the infected internal organs were not correlated with mortality but some degenerative changes were observed in all the infected organs within a week. Whitish nodules in the infected organs which are the typical character in nocardial infection was initially found at two weeks after the artificial infection in snakehead.

• Journal of the HCI Society of Korea
• /
• v.13 no.3
• /
• pp.65-73
• /
• 2018

Game balance settings are crucial to game design. Game balancing must take into account a large amount of numerical values, configuration data, and the relationship between elements. Once released and served, a game - even for a balanced game - often requires calibration according to the game player's preference. To achieve sustainability, game balance needs adjustment while allowing for small changes. In fact, from the producers' standpoint, game balance issue is a critical success factor in game production. Therefore, they often invest much time and capital in game design. However, if such a costly game cannot provide players with an appropriate level of difficulty, the game is more likely to fail. On the contrary, if the game successfully identifies the game players' propensity and performs self-balancing to provide appropriate difficulty levels, this will significantly reduce the likelihood of game failure, while at the same time increasing the lifecycle of the game. Accordingly, if a novel technology for game balancing is developed using artificial intelligence (AI) that offers personalized, intelligent, and customized service to individual game players, it would bring significant changes to the game production system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.774-784
• /
• 2021

The object detection and recognition process is a very important task in the field of computer vision, and related research is actively being conducted. However, in the actual object recognition process, the recognition accuracy is often degraded due to the resolution mismatch between the training image data and the test image data. To solve this problem, in this paper, we designed and developed an integrated object recognition and super-resolution framework by proposing an image super-resolution technique to improve object recognition accuracy. In detail, 11,231 license plate training images were built by ourselves through web-crawling and artificial-data-generation, and the image super-resolution artificial neural network was trained by defining an objective function to be robust to the image flip. To verify the performance of the proposed algorithm, we experimented with the trained image super-resolution and recognition on 1,999 test images, and it was confirmed that the proposed super-resolution technique has the effect of improving the accuracy of character recognition.