• 방송공학회논문지
• /
• 제25권4호
• /
• pp.620-628
• /
• 2020

하이라이트를 자동으로 예측 하는 문제는 영상을 사람이 직접 편집하는 시간과 비용 문제를 해결하기 위해 필요한 기술이다. 본 논문에서는 하이라이트 구간 내에서 하이라이트 판단 여부에 영향을 주는 특정 부분에 집중하기 위해 앙상블 모델을 제안한다. 우리의 모델은 하나의 단일 모델만으로는 충분히 학습하기 어려운 중요한 정보를 앙상블을 통해 더 많은 유용한 특징들을 얻을 수 있다. 앙상블을 이루는 단일모델들은 오디오와 이미지 정보를 결합하여 다양한 영상의 특징들을 추출한다. 직접 수집한 e스포츠 경기 영상과 야구 경기 영상을 통해 하이라이트 예측 성능이 개선됨을 확인한다.

• 한국의류산업학회지
• /
• 제10권2호
• /
• pp.138-146
• /
• 2008

Opera costumes are a tangible medium to directly affect how opera singers performed. Therefore, stage costumes have been designed in various ways to reflect intentions of directors. This study examines modernization of opera costumes. For this study, the author worked as one of costume design staff for "Faust" which was played in Sung-Nam Art Center from November 24, 2005 to November 27, 2005 in commemoration of the opening of the center. The following outcomes of the supporting actors are drawn from this study. Martha wore a grey jacket and a long skirt with slash. Grey suit symbolized chastity but tight long skirt with slash means sexy. Valentin wore military and combat uniforms to express that he was a soldier. The military uniform represented the situation before the war and the combat uniform symbolized the situation after the war. Rather short trousers, a white tweed jacket, purple flower ornament, a light green vest, and a beret were used to express Wagner, who looked ridiculous and always tried to flirt. Overalls, a blue jumper, and a scarlet baseball cap were used to make Siebel look naive and homely. Modern opera costumes now help reduce the time and space between stage and audience, improve economic efficiency and meet the audience's needs for various style.

• Journal of Nutrition and Health
• /
• 제37권5호
• /
• pp.364-372
• /
• 2004

The purpose of this study was to investigate the possibility of medicinal plants application as an edible functional food resource. We carried out to develop a traditional functional beverage by using hot-water extraction of 4 medicinal plants (Polygonatun sibiricum, Ophiopogonis radix, Lycii fructus, Schizandriae frutcus) and we examined the effects of drink on physiological response during exercise and recovery phase. The subjects were male baseball players, and exercise protocol was performed with 45 minutes treadmill running and 85％ VO$_2$ max intensity. Brix, pH and titratable acidity of developed drink were 9.5％, 3.3 and 0.22％. The approximate nutritional composition of beverage was carbohydrate, 5.98％, total dietary fiber, 0.42％, crude protein, 0.70％, crude fat, 0.20％ and crude ash, 0.20％. Developed. traditional functional beverage contained K (4.00 mg％), Na (3.68mg％), Ca (2.54mg％), Mg (1.60mg％) and Fe (0.29mg％). Developed beverage drinking group showed a lowest heart rate during exercise and recovery phase than the other two group (water group and ion beverage group). In the change of blood lactate concentration, developed beverage intake group showed a significant lowest values during exercise and recovery phase. And blood hematocrit values and osmolarity were lowest in the developed beverage intake group. In these results, exercise performance capacity was expected to improve most highly with the developed drink from composition with medicinal plants. Also developed beverage was effective in recovery of exercise-induced fatigue. Thus developed drink with medicinal plants can be used as a functional material improving decrease fatigue effects in beverage industry.

• 대한안전경영과학회지
• /
• 제15권2호
• /
• pp.255-262
• /
• 2013

Due to the competition between the various professional events, it is imperative for the team's management to improve efficiency by removing the inefficiencies of the professional team in order to gain a competitive edge. This study use different Data Envelopment Analysis (DEA) models to measure the efficiency of professional sport organizations. In this mathematical-analytical study, this study first reviews the related literature to analyze the input/output variables. In the end, the variables were detected and the data for this study were gathered from the Korean Basketball League (KBL). While previous studies examine relative efficiency of Korean Professional baseball teams by using CCR model, this study fully utilize the DEA method to investigate Korean professional sports organizations' operating problem. Thus, this study propose full results of DEA analysis such as efficiency score (overall, technical, and scale efficiency), slacks in inputs and outputs of inefficient organizations, Malmquist index) As a result, this study provides not only the exact productivity information of a team and a way of improving a firm's productivity with a decision maker.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(4)
• /
• pp.5-8
• /
• 2001

We propose a realtime numeric caption recognition algorithm that automatically recognizes the numeric caption generated by computer graphics (CG) and displays the modified caption using the recognized resource only when a valuable numeric caption appears in the aimed specific region of the live sportscast scene produced by other broadcasting stations. We extract the mesh feature from the enhanced binary image as a feature vector after acquiring the sports broadcast scenes using a frame grabber in realtime and then recover the valuable resource from just a numeric image by perceiving the character using the neural network. Finally, the result is verified by the knowledge-based rule set designed for more stable and reliable output and is displayed on a screen as the converted CC caption serving our purpose. At present, we have actually provided the realtime automatic mile-to-kilometer caption conversion system taking up our algorithm f3r the regular Major League Baseball (MLB) program being broadcasted live throughout Korea over our nationwide network. This caption conversion system is able to automatically convert the caption in mile universally used in the United States into that in kilometer in realtime, which is familiar to almost Koreans, and makes us get a favorable criticism from the TV audience.

• 산업경영시스템학회지
• /
• 제25권3호
• /
• pp.16-27
• /
• 2002

This paper is mainly interested in customer's(fan's) royalty to professional sports in sports industry. The royalty is studied with the royalty to several products and a service, so that we can obtain the comparison between both the former and the latter royalty. Interested professional sports, products and service are baseball, basketball, soccer, beer, cu, cola/cider, etc, and we will call each of them 'item'. The seven related hypotheses are considered. In hypotheses the dependent variable is royalty to nine items, and the independent variables are rune items, and customer's sex and major. For testing hypotheses, a questionaire is used, and reliability is checked with Cronbach's Alpha and correlation coefficients. For some items the royalty is significantly strong. Rather, even if some item has bad quality, it is found that customers can be royal to that item. We can identify some of those items through the results of this study. The strong significant results will be useful for the customer management. Additional studies are required for nonsignificant results which seems to be useful, and they are noted in the final part of the paper.

• 구강회복응용과학지
• /
• 제16권3호
• /
• pp.229-236
• /
• 2000

The purpose of this study was to analyze the prevalence of Oral and maxillofacial injuries of elementary school children in Suwon, Korea. The sample consisted of 850 boys and girls in a elementary school children in the city. This study conducted a survey by sending questionnaires to school-aged boys and girls, to study a condition of traumatic injuries of teeth. 1. The students who had a traumatic injuries in a oral & maxillofacial area are 125 persons(14.7%); boys 93 persons(21%), girls 32 persons (7.9%). 2. Incidence of traumatic injuries was 1 time 102 persons(81.6%), 2 times 14 persons(11.2%), others 9 persons(7.2%). 3. The causes of traumatic injuries were fall down 65.6%, blow 10.4%, sports 9.6%, traffic accidents 7.2%, tripping 3.2%, others 4%. 4. The distributions of sports related traumatic injuries were roller blade 7 persons, bicycle 2 persons, football 1 person, baseball 1 person, basketball 1 person. 5. Types of traumatic injuries were soft tissue injury(40.8%), tooth fracture(36.8%), extrusion(8.0%), jaw fracture(3.2%), others(11.2%).

• Journal of the Korean Data and Information Science Society
• /
• 제25권3호
• /
• pp.493-499
• /
• 2014

야구의 승률은 총득점의 제곱을 총득점의 제곱과 총실점의 제곱의 합으로 나눈 것으로 추정된다는 야구의 피타고라스 정리에 대하여 많은 연구들이 활발하게 진행되고 있다. 본 연구에서는 피타고라스 정리에 사용되는 지수에 대한 새로운 추정방법을 제안하며 평균제곱오차의 제곱근 (root mean squared error; RMSE)을 이용하여 널리 알려진 추정방법들과 상대적 효율성을 비교하였다. 사용된 데이터는 1982년부터 2013년 사이의 모든 한국프로야구 기록이며, 그 결과 제안된 방법은 기존의 방법보다 RMSE 관점에서 바람직하다고 간주된다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2014년도 추계학술발표대회
• /
• pp.721-722
• /
• 2014

야구는 흔히 기록의 스포츠라는 별명으로 많이 불린다. 그만큼 야구라는 운동이 갖는 기록의 종류는 무척 다양하고 또한 기록의 활용 가능성 역시 무궁무진하다. 이러한 별명에 걸맞게 미국에서는 야구에 대한 다양하고 방대한 정보를 수집하고 활용하고 있다. 그러나 한국 프로야구에 대한 정보의 수집과 활용은 아직까지 크게 부각되지 못하는 것이 현실이다. 랜덤 포레스트 기법을 이용하여 경기의 승부를 예측함으로써 한국 프로야구 데이터의 수집과 활용을 증대 시키는 효과를 기대 해 본다. 본 논문에서는 2014년 한국 프로야구의 승부 예측을 주제로 어떠한 누적 스포츠 데이터집단이 가장 유효한지를 실험 하였다. 승부 예측을 하기위해 사용된 누적 스포츠 데이터는 2014년 선수와 팀 기록, 2013부터 2014년까지의 선수와 팀 기록, 2012년부터 2014년까지의 선수와 팀 기록이다. 이들 세 그룹의 데이터를 이용하여 이분데이터 모형에 랜덤 포레스트 기법을 사용한 승부예측 알고리즘에 적용 시킨 후 어느 그룹의 데이터가 가장 실제 2014 한국 프로야구 정기결과와 맞을 확률이 높음을 구하여 가장 유용한 데이터 그룹이 어떤 그룹인지 연구 하였다.

• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1564-1573
• /
• 2016

With the spread of social media and mobile devices, people spend more time on online than ever before. As more people participate in various online activities, much research has been conducted on how to make use of the time effectively and productively. In this paper, we propose two methods which can be used to extract highlights and make searchable media indexes using online social data. For highlight extraction, we collected the comments from the online baseball broadcasting website. We adopted peak-finding algorithm to analyze the frequency of comments uploaded on the comments section of the website. For each indexes, we collected postings from soap opera forums provided by a popular web service called DCInside. We extracted all the instances when a character's name is mentioned in postings users upload after watching TV, which can be used to create indexes when the character appears on screen for the given episode of the soap opera The evaluation results shows the possibility of the crowdsourcing-based media interaction for both highlight extraction and index building.