• 국제학술발표논문집
• /
• The 10th International Conference on Construction Engineering and Project Management
• /
• pp.487-493
• /
• 2024

Presently, prefabricated concrete panels are extensively employed in diverse construction projects across the globe due to their exceptional quality. To maintain the overall quality of these construction projects, it is crucial to ensure that the dimensions of precast concrete panels align with their designated design specifications. Therefore, it is essential to develop a methodology capable of quickly and accurately measuring the dimensions of precast concrete panels. Currently, there are many advanced technologies used to examine the dimensions of prefabricated concrete panels such as terrestrial laser scanning, which is prone to time consuming and cost inefficiencies. To address these limitations, this study suggests a computer vision-based approach that utilizes April Tag markers and images taken from a mobile phone to measure and evaluate the dimensions and quality of precast concrete panels. The proposed algorithm operates as follows: Initially, the RGB image coordinates are converted to the world coordinate systems using April tag markers. Following, the masks of the precast concrete components are extracted using the state-of-the-art Segment Anything Model (SAM). Finally, an algorithm based on image processing technique is developed to estimate the dimensional properties of precast concrete panels. The effectiveness of the proposed method is validated through preliminary experiments conducted in the field-scale precast slabs, and the result is evaluated by comparing to the manual measurement result.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.344-352
• /
• 2022

Accurate indoor localization of construction workers and mobile assets is essential in safety management. Existing positioning methods based on GPS, wireless, vision, or sensor based RTLS are erroneous or expensive in large-scale indoor environments. Tightly coupled sensor fusion mitigates these limitations. This research paper proposes a state-of-the-art positioning methodology, addressing the existing limitations, by integrating Stereo Visual Inertial Odometry (SVIO) with fiducial landmarks called AprilTags. SVIO determines the relative position of the moving assets or workers from the initial starting point. This relative position is transformed to an absolute position when AprilTag placed at various entry points is decoded. The proposed solution is tested on the NVIDIA ISAAC SIM virtual environment, where the trajectory of the indoor moving forklift is estimated. The results show accurate localization of the moving asset within any indoor or underground environment. The system can be utilized in various use cases to increase productivity and improve safety at construction sites, contributing towards 1) indoor monitoring of man machinery coactivity for collision avoidance and 2) precise real-time knowledge of who is doing what and where.

• 수산해양기술연구
• /
• 제51권2호
• /
• pp.227-234
• /
• 2015

To obtain geographical range and growth-kinetics parameters of mottled sake (Beringraja pulchra) populations in the Yellow Sea, three mark-recapture experiments were carried out. Overall, 991 tagged individuals were released, and 4.1% of them were recaptured with the mean release period of 339 d (range, 8-1,420 d) and the mean growth rate of $1.4cm\;mon^{-1}$ (female, $1.5cm\;mon^{-1}$; male $1.3cm\;mon^{-1}$). In the first experiment, 667 individuals were released at Heuksan Island from April to June, 2007-2009, and 30 individuals were recaptured mainly at the north and the north-east coasts of the island, indicating absence of migration to the south of the island. In the second experiment, 323 individuals were released at several fishing grounds scattered in the Yellow Sea in 2010-2013, and 11 individuals were recaptured at points deviated to all directions from the releasing points. As the last, one individual was released with pop-up satellite archival tag at a costal point ($34^{\circ}37.2$'N, $124^{\circ}59.3$'E) off Hong Island on May 21, 2010. The tagged individual migrated to a north-east location ($35^{\circ}50.4$'N, $126^{\circ}03.6$'E) of Eocheong Island by Aug. 25, 2010. The data archived for the three months in the tag indicated that the migration path had depths of 48-80 m and temperature of $12.6-14.4^{\circ}C$. The results indicated that mottled sake populations had a localized habitat ranges at the north of Heuksan Island and the west of Hong Island while growing at the rate of $1.4cm\;mon^{-1}$.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2021년도 추계학술대회
• /
• pp.9-12
• /
• 2021

드론 Photogrammetry는 높은 기술적 활용가치가 있는 기술로서, 결과물로 생성하는 3D 디지털 공간정보 모델이 시설물의 비육안 안전점검 및 진단에 활용될 수 있을 뿐만 아니라 디지털 트윈 구축을 위한 가장 기초적이고 핵심적인 수치 데이터를 제공하기 때문이다. 본 연구에서는 드론 Photogrammetry의 적정 품질을 구현하기 위한 임무비행의 다양한 영향인자에 대해 고찰하였다. K-water연구원 누수탐사실습장을 대상으로 드론 사진 촬영 시 비행고도, 비행속도, 중첩도, 카메라 Pitch각의 영향에 대해 연구를 수행하였다. 본 연구에서 비행시간에 영향을 미치는 인자로서 비행고도, 중첩도, 비행속도의 순으로 중요도가 있음을 알 수 있었다. 드론 임무 비행 시 후처리 결과에 가장 큰 영향을 미치는 인자는 중첩도로 나타났다. 중첩도 60% 임무비행은 3D 모델의 geometry 왜곡이 큰 편으로 나타났다. 비행 고도는 GSD (Ground Sampling Distance)와 직접 연계되므로 중요하며, 낮은 고도일수록 높은 품질의 모델링이 가능하다. April Tag를 통한 지상기준점 자동 패턴 인식 기능은 후처리 과정에서 시간 절약이 가능하여 유용하였다. 비행속도에 의한 결과물의 품질은 큰 차이가 없었으나, 수직 구조물의 모서리 부분에 다소 차이가 있었다. 짐벌 Pitch각도에 의한 정사영상 품질의 차이는 크지 않았으나 수직구조물과 평면적 구조물에 따라 각기 다른 촬영각도를 적용하는 것이 바람직하다. 본 연구성과는 향후 보다 다양한 환경에서의 데이터 수집을 통해 최적 디지털 현실 모델링에 기여할 것으로 판단된다.

• 한국어류학회지
• /
• 제22권2호
• /
• pp.83-89
• /
• 2010

본 연구는 우리나라 남해안에 서식하는 문치가자미(pleuronectes yokohamae)의 성숙과 산란에 관한 연구로, 2006년 5월부터 2007년 4월까지 경남 남해군 연안 호망어업과 연안자망에서 어획된 것을 무작위로 채집하여 조사하였다. 암컷의 생식소 발달과정 및 생식소 중량지수의 월 변화로부터 문치가자미의 산란기는 11~2월, 주 산란기는 12~1월로 추정된다. 전장(TL)과 포란수(F)간의 관계식은 $F=3.656TL^{3.518}$ ($R^2$=0.872)이며, 체장이 커질수록 포란수도 증가하는 경향을 보였다. 산란에 참여하는 50% 성숙체장은 전장 19.8 cm이었다. 전 조사기간에 대한 성비(암 수)는 55.2:44.8로 암컷이 우세하였고, 체급별 성비는 전장 25 cm 이하에서는 수컷의 비율이 높았으나 체장이 커질수록 암컷의 비율이 증가하여 32 cm 이상에서는 전 출현개체수가 암컷이었다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제19권4호
• /
• pp.593-600
• /
• 2006

The effects of breed and sex on individual growth performance and feeding behaviour were studied between 45 and 90 kg BW in two replicates of forty group-housed pigs. The first and the second replicates were carried out during the warm season (i.e. between February and April 2003) and during the hot season (i.e. between August and October 2003), respectively. During the warm season, ambient temperature and relative humidity averaged $25.3^{\circ}C$ and 86.0%. The corresponding values for the hot season were $27.9^{\circ}C$ and 83.6%. The pigs were grouped in pens of 10 animals on the basis of breed (Creole or Large White) and sex (gilt or castrated male) and given ad libitum access to a grower diet (9.0 MJ/kg net energy and 158 g/kg crude protein) via feed intake recording equipment (Acema 48). An ear-tag transponder was inserted into each pig and this allowed the time, duration, and size of individual visits to be recorded. The growth performance and feeding pattern were significantly affected by breed, sex, and season. The Creole pigs (CR) had a lower average daily gain (ADG) (642 vs. 861 g/d, p<0.01) and carcass lean content ($LC_{90kg}$) (35.4 vs. 54.5%; p<0.01) and a higher backfat thickness at 90 kg BW ($BT_{90kg}$) (23.4 vs. 10.4 mm; p<0.01) than Large White pigs (LW) whereas the average daily feed intake (ADFI) was not affected by breed (2.34 vs. 2.22 kg/d, respectively for CR and LW pigs; p>0.10). Consequently, the food:gain ratio was higher in CR than in LW (3.65 vs. 2.58; p<0.01). CR had less frequent meals but ate more feed per meal than LW (5.9 vs. 8.8 meals/d and 431 vs. 279 g/meal; p<0.01). The rate of feed intake was lower (27.6 vs. 33.9 g/min; p<0.01) and the ingestion time per day and per meal were higher in CR than in LW (87.1 vs. 69.7 min/d and 15.8 vs. 8.4 min/meal; p<0.01). The ADFI and BT90 kg were higher (2.38 vs. 2.17 kg/d and 18.1 vs. 15.9 mm; p<0.05) and LC90 kg was lower (43.5 vs. 46.4%; p<0.01) in castrated males (CM) than in gilts (G) whereas ADG was not affected by sex (p = 0.12). The difference in lean content between CM and G was greater in CR than in LW. The ADFI and ADG were reduced during the hot season (2.18 vs.2.38 kg/d and 726 vs. 777 g/d, respectively; p<0.05) whereas feed conversion and carcass lean content were not affected by season (p>0.05). Average feeding time per meal and meal size decreased during the hot season (10.9 vs. 13.2 min/meal and 316 vs. 396 g/meal; p<0.01) whereas the rate of feed intake was not affected by season (p = 0.83). On average, 0.69 of total feed intake was consumed during the diurnal period. However, this partition of feed intake was significantly affected by breed, sex, and season. In conclusion, the breed, sex and season significantly affect performance and feeding pattern in growing pigs raised in a tropical climate. Moreover, the results obtained in the present study suggest that differences observed in BW composition between CR and LW are associated with difference in feeding behaviour, in particular, the short-term regulation of feed intake.