http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
음향 이벤트 검출 모델에 적용된 Pooling 방법에 따른 성능 변화에 관한 고찰
홍성재(Sungjae Hong),박상욱(Sangwook Park) 대한전자공학회 2024 대한전자공학회 학술대회 Vol.2024 No.6
This study explored the effectiveness of combining max pooling and average pooling techniques to enhance the performance of acoustic event recognition systems. Through experiments conducted using the DESED dataset, it was found that the model employing max pooling at the initial layer significantly droped recognition performance compared to the baseline model. This finding suggests that there may be an optimized pooling technique suitable for each layer within the model. Future work will focus on further enhancing performance by developing and applying layer-specific optimized pooling techniques.
8주간 SAQ 트레이닝이 신체특성과 근 활성도에 미치는 영향
장창호(ChangHo Jang),홍성재(SungJae Hong),박준성(Joonsung Park),심힘찬(HimChan Shim),김종빈(JongBin Kim) 한국콘텐츠학회 2023 한국콘텐츠학회논문지 Vol.23 No.2
본 연구는 8주간 SAQ(Speed, Agility, Quickness)트레이닝을 통해 신체조성, 하지 근 활성도를 입증하는데 목적이 있다. 20대 남성 24명(나이:22.78±2.28 year, 신장:1.74±0.06 m, 체중:71.61±10.38 kg)을 대상으로 SAQ트레이닝 동작을 최대 반응 속도의 강도로 훈련을 실시하였다. 소요 시간 및 하지 근 활성도을 이원반복분산분석을 실시하였다. Knee up 소요시간은(p<.009), Reaction Time 소요시간(p<.018), 20m 달리기 시기(p<.05), 근 활성도는 side step 시 왼쪽 비복근 집단(p<.038), 오른쪽 대퇴이두근(p<.002),왼쪽 대퇴이두근(p<.001), 오른쪽 대퇴직근(p<.001), 오른쪽 대퇴이두근(p<.001)과 왼쪽 대퇴직근(p<.001)에서 유의한 차이가 나타났으며, Reaction Time에서는 왼쪽 비복근(p<.001)과 오른쪽 대퇴이두근(p<.001)은 시기에서 차이가 나타났다. SAQ트레이닝 시 민첩성, 스피드, 경기력 및 체력향상으로 트레이닝 방법론의 학술적인 자료로 활용될 것으로 사료된다. The purpose of this study is to demonstrate body composition and lower extremity muscle activity through SAQ (Speed, Agility, Quickness) training for 8 weeks. 24 men in their 20s (age: 22.78±2.28 years, height: 1.74±0.06 m, weight: 71.61±10.38 kg) were trained with SAQ training at the maximum response speed intensity. Two-way repeated analysis of variance was performed on the required time and lower extremity muscle activity. Knee up time required (p<.009), reaction time time required (p<.018), 20m running time (p<.05), muscle activity during side step left gastrocnemius group (p<.038), right Significant in biceps femoris (p<.002), left biceps femoris (p<.001), right rectus femoris (p<.001), right biceps femoris (p<.001) and left rectus femoris (p<.001) There was a difference, and in the reaction time, the left gastrocnemius (p<.001) and the right biceps femoris (p<.001) showed a difference in timing. It is considered that it will be used as an academic data for training methodology due to the improvement of agility, speed, performance and physical strength during SAQ training.
정지궤도 위성 대류권 오존 관측 자료를 이용한 대류권 이동벡터 산출 가능성 연구
신대근,김소명,박주선,백강현,홍성재,김재환,Shin, Daegeun,Kim, Somyoung,Bak, Juseon,Baek, Kanghyun,Hong, Sungjae,Kim, Jaehwan 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6
The tropospheric ozone is a pollutant that causes a great deal of damage to humans and ecosystems worldwide. In the event that ozone moves downwind from its source, a localized problem becomes a regional and global problem. To enhance ozone monitoring efficiency, geostationary satellites with continuous diurnal observations have been developed. The objective of this study is to derive the Tropospheric Ozone Movement Vector (TOMV) by employing continuous observations of tropospheric ozone from geostationary satellites for the first time in the world. In the absence of Geostationary Environmental Monitoring Satellite (GEMS) tropospheric ozone observation data, the GEOS-Chem model calculated values were used as synthetic data. Comparing TOMV with GEOS-Chem, the TOMV algorithm overestimated wind speed, but it correctly calculated wind direction represented by pollution movement. The ozone influx can also be calculated using the calculated ozone movement speed and direction multiplied by the observed ozone concentration. As an alternative to a backward trajectory method, this approach will provide better forecasting and analysis by monitoring tropospheric ozone inflow characteristics on a continuous basis. However, if the boundary of the ozone distribution is unclear, motion detection may not be accurate. In spite of this, the TOMV method may prove useful for monitoring and forecasting pollution based on geostationary environmental satellites in the future.