• 한국산림과학회지
• /
• 제65권1호
• /
• pp.96-99
• /
• 1984

6 년생(年生) 잣나무 조림지(造林地)에 있어서 임분(林分)의 대표치(代表値) 측정(測定)을 위한 적정(適定) 플롯의 크기를 조사분석(調査分析)하였는 바, 수고(樹高)에 대하여는 0.04ha 크기의 플롯이, 흉고직경(胸高直徑)에 대하여는 0.08ha 크기의 플롯이 적합하였다.

• 한국의류산업학회지
• /
• 제7권1호
• /
• pp.4-12
• /
• 2005

The purpose of this study was to analyze the utilization of Fashion Brands Web-site. This analysis was done by 6C Concept(Contents, Community, Commerce, Connection, Customizing, Communication) which was Internet Marketing Strategy. The review of previous studies and empirical investigations through the Internet were processed for this study. 151 fashion brands in department stores in Daegu, Korea were surveyed from January to February 2004. Data were analyzed by using frequency and percentage. Total 105 brands established their Web-Sites of Internet out of 151(69.5%) fashion brands. There were four characters on Contents, six characters on Connection and five characters on Communication. And there were a establishing rate of 38.1 percent on Online Community and 30.5 percent on Online Shopping Mall. On Customizing, 73.3 percent of brands had e-CRM systems.

• 대한인간공학회지
• /
• 제6권2호
• /
• pp.9-18
• /
• 1987

The purpose of this study was to find the optimal stroke width-to-height ratio of the Korean character (Hangeul) which yields the greatest reading distance. In this study, black characters on a white background were used under 200 lux conditions in-doors, with the stroke width-to-height ratios ranging from a very thin 1:125 to heavier 1: 8.3. The criterion was the average reading distance at which three subjects (male university students) having normal visual acuity (1.2) could read the characters. Using the data obtained from the experi- ment, we analyzed the relationship between stroke width-to-height ratio and reading distance. In the case of the black characters on a white background, the greatest reading distances occurred with a range from 1:11.4 to 1:17.9.

• International journal of advanced smart convergence
• /
• 제11권3호
• /
• pp.150-156
• /
• 2022

Recently, the use of Unreal Engine for animation production is increasing. In this situation, Unreal Engine's 'MetaHuman Creator' helps make it easier to apply realistic characters to animation. In this regard, we tried to produce animations using 'MetaHuman' and verify the effectiveness and differences from the animation production process using only Maya software. To increase the efficiency of the production process, the animation process was made with Maya software. We tried to import animation data from Unreal Engine and go through the process of making animations, and try to find out if there are any problems. And we tried to compare animations made with realistic 'MetaHuman' characters and animation works using cartoon-type characters. The use of the same camera lens in realistic character animations and cartoon character animations produced based on the same scenario was judged to be the cause of the lack of realistic animation screen composition. The analysis revealed that a different approach from the existing animation camera lens selection is required for the selection of the camera lens in the production of realistic animation.

• International Journal of Industrial Entomology and Biomaterials
• /
• 제19권1호
• /
• pp.205-213
• /
• 2009

In order for further systematic maintenance of silkworm stocks kept in Korea we analyzed character quality of a diverse array of silkworm strains originated from several sericulture-practicing countries. The analysis of about ten qualitative characters from 67 strains (13 of Japanese strains, 15 of Chinese strains, 14 of European strains, 6 of Korean and Tropical strains, and 19 of unknown origin) revealed a significant difference in the ten different qualitative characters among silkworm strains. In the analysis of quantitative characters, Japanese and European strains were highest in hatchability, the Korean and Tropical strains were highest in pupation rate, and unknown origin and Chinese strains were highest in cocoon yield and number of egg laid. With the connection of molecular genetic analysis the current data may provide the advanced ground for further systematic maintenance of valuable genetic resources of silkworms, although more breeds should be investigated for further complete pictures.

• 한국멀티미디어학회논문지
• /
• 제23권11호
• /
• pp.1361-1371
• /
• 2020

Fonts have developed as a visual element, and their influence has rapidly increased around the world. Research on font automation is actively being conducted mainly in English because Hangul is a combination character and the structure is complicated. In the previous study to solve this problem, the stroke element of the character was automatically extracted by applying the object detection by component. However, the previous research was only for similarity, so it was tested on various print style fonts, but it has not been tested on other characters. In order to extract the stroke elements of all characters and fonts, we performed a performance analysis experiment according to the expansion character in the Hangul stroke element extraction training. The results were all high overall. In particular, in the font expansion type, the extraction success rate was high regardless of having done the training or not. In the character expansion type, the extraction success rate of trained characters was slightly higher than that of untrained characters. In conclusion, for the perfect Hangul stroke element extraction model, we will introduce Semi-Supervised Learning to increase the number of data and strengthen it.

• 한국컴퓨터정보학회논문지
• /
• 제25권7호
• /
• pp.39-46
• /
• 2020

최근, 대전 게임 분야에서, 가상 캐릭터들의 효과적인 설계를 위해 캐릭터의 승률을 효율적으로 예측할 수 있는 방법들이 매우 필요하다. 우리는 본 논문에서 이 문제를 해결하기 위해 시뮬레이션과 기계 학습을 결합하는 방법을 제안한다. 우선 대전 게임에서 가상 캐릭터의 대전 승률을 분석하기 위해서 시뮬레이션을 사용하고, 가상 캐릭터의 능력치에 따라서 승률을 예측하기 위해 회귀 모델에 기반한 기계 학습 기법을 적용한다. 제안한 기법으로 실험한 결과는 시뮬레이션 결과로 나온 승률과 기계 학습 기법이 예측한 승률이 거의 차이가 없다는 것을 확인하였다. 그리고 간단한 회귀 모델에 기반한 기계 학습으로도 실험에서 좋은 성능을 얻을 수 있었다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제15권2호
• /
• pp.9-17
• /
• 2009

다수 캐릭터가 경쟁적으로 상호작용하는 애니메이션의 합성은 컴퓨터 게임, 애니메이션 등의 응용분야에서 종종 요구되는 중요한 문제이다. 하지만 상대의 예측하기 어려운 행동에 효과적으로 대응하는 전략적 경쟁 양상을 모사하는 것은 어려운 문제로 남아있다. 본 논문은 다수 에이전트 학습 분야에서 제안된 마르코프 게임 강화학습 알고리즘을 촬영된 동작 데이터로부터 생성된 행위 모델에 적용하여 사실적인 경쟁 애니메이션을 합성하는 방식을 제안한다. 추격-회피, 간격 유지, 총격전 등의 다양한 경쟁적 상황에 대하여 효과적인 전략을 학습하여 흥미로운 애니메이션을 합성하는 예제들을 통하여 본 논문이 제안하는 방법의 효용성을 보인다.

• Journal of Information Technology Applications and Management
• /
• 제31권4호
• /
• pp.149-162
• /
• 2024

This paper explored the characters in spin-off works that are becoming a hot topic not only in Korea but also in the world. I would like to find out whether there are changes in the character in the original and the character's character and behavior in the spin-off work, and if there are changes, for what reason the changes occur. The subjects of the study were "Puss in Boots" and "Minions," which released a number of spin-off works based on the original. As analysis tools, we used the five-step theory of need by Abraham H. Maslow and the behavioral model of Greimas. As a result, in the spin-off work, Minions and Puss have shifted away from the role of facilitator in the original and become subjective characters. According to Maslow's theory of needs, the analysis also confirmed that this appears as an action to satisfy the needs of belonging, love, respect, and self-realization beyond basic physiological and safety needs. This change allowed him to develop a more independent and proactive personality and behavior, and be at the center of the story. This is the result of reflecting the production team's intention and audience expectations to provide audiences with deeper characters and richer stories. is analyzed.

• 한국통신학회논문지
• /
• 제21권12호
• /
• pp.3265-3276
• /
• 1996

문자인식에 대한 연구는 주로 한글인식에 대해서만 이루어져 왔는데, 대부분의 문서는 한글 뿐만 아니라 여러 종류의 문자가 포함되어 있다. 따라서, 본 논문에서는 다중 크기, 다중 활자체, 다자종 문자가 포함되어 있는 한글문서를 인식할 수 있는 문자인식 시스템을 구현하였다. CombNET 구조를 갖는 신경회로망을 자종별로 구성하여, 문자인식시에 문자를 구별하지 않고 인식하는 방법을 제안하였다. CombNET 구조의 상단부를 차지하는 Kohonen의 SOFM 신경망을 이용하여 한글과 한자는 36개, 영숫자는 16개의 유형으로 분류하고 각 유형에 대해서 CombNET 구조의 하단부에 있는 BP 네트워크를 이용하여 문자인식을 수행하였다. 실험결과 학습 데이타에 대해서는 95.6%의 인식율을 나타내었고, 실제문서에 대해서도 92.6%의 인식율과 초당 10.3자의 인식속도를 보임으로써 제안된 인식 시스템의 유효성을 입증하였다.