• Journal of Korean Tunnelling and Underground Space Association
• /
• v.26 no.2
• /
• pp.129-152
• /
• 2024

In the application of deep learning object detection via CCTV in tunnels, a large number of false positive detections occur due to the poor environmental conditions of tunnels, such as low illumination and severe perspective effect. This problem directly impacts the reliability of the tunnel CCTV-based accident detection system reliant on object detection performance. Hence, it is necessary to reduce the number of false positive detections while also enhancing the number of true positive detections. Based on a deep learning object detection model, this paper proposes a false positive data training method that not only reduces false positives but also improves true positive detection performance through retraining of false positive data. This paper's false positive data training method is based on the following steps: initial training of a training dataset - inference of a validation dataset - correction of false positive data and dataset composition - addition to the training dataset and retraining. In this paper, experiments were conducted to verify the performance of this method. First, the optimal hyperparameters of the deep learning object detection model to be applied in this experiment were determined through previous experiments. Then, in this experiment, training image format was determined, and experiments were conducted sequentially to check the long-term performance improvement through retraining of repeated false detection datasets. As a result, in the first experiment, it was found that the inclusion of the background in the inferred image was more advantageous for object detection performance than the removal of the background excluding the object. In the second experiment, it was found that retraining by accumulating false positives from each level of retraining was more advantageous than retraining independently for each level of retraining in terms of continuous improvement of object detection performance. After retraining the false positive data with the method determined in the two experiments, the car object class showed excellent inference performance with an AP value of 0.95 or higher after the first retraining, and by the fifth retraining, the inference performance was improved by about 1.06 times compared to the initial inference. And the person object class continued to improve its inference performance as retraining progressed, and by the 18th retraining, it showed that it could self-improve its inference performance by more than 2.3 times compared to the initial inference.

• Journal of the Korean Applied Science and Technology
• /
• v.41 no.2
• /
• pp.219-232
• /
• 2024

The aim of this study was to explore the impact of body mass index (BMI) and drumming exercise on autonomic nervous system function in adult women. Ten adult women (aged 30-50) were divided into two groups based on their BMI: a normal BMI group (Low BMI, LBMI < 23 kg/m²) and an overweight BMI group (High BMI, HBMI > 23 kg/m²). Both groups participated in a drumming exercise program, consisting of 50-minute sessions, three times a week, for a duration of 8 weeks. Body composition and heart rate variability were assessed before and after the 8-week exercise period. Heart rate variability was evaluated using linear analysis in the time domain and frequency domain. Additionally, non-linear analysis was conducted using a Poincaré plot. The autonomic nervous system index was determined by measuring parasympathetic nervous system index and sympathetic nervous system index. Following the 8-week intervention, the HBMI group exhibited a significant decrease in weight (p=0.034), BMI (p=0.044), body fat mass (p=0.032), and waist circumference (p=0.013) compared to the LBMI group. Furthermore, the HBMI group demonstrated significant increases in RMSSD (p=0.018) and TP (p=0.033) in linear analysis, as well as SD1 (p=0.018) in non-linear analysis and PNS Index (p=0.040) compared to the LBMI group. RMSSD, SD1, and PNS Index serve as indicators of parasympathetic nervous system activity in linear and non-linear analyses, respectively. These findings indicate that drumming exercise significantly enhances autonomic nervous system function in overweight women.

• Journal of Food Hygiene and Safety
• /
• v.39 no.3
• /
• pp.288-298
• /
• 2024

This study aimed to investigate the amino acid content, fatty acid composition, total flavonoid and phenol contents, and antioxidant activities of Prunus yedoensis Matsum. bark extracts using various extraction solvents. Through amino acid analysis, 13 amino acids were detected in extracts obtained using hot water, 30% ethanol, and 70% ethanol. The major amino acids were identified as aspartic acid, arginine, and proline, and the total amino acid content was 0.17%, 0.16%, and 0.09%, respectively. Fatty acid analysis showed a saturated fatty acid (SFA) ratio of 62.7-66.7% in extracts obtained using hot water, 30% ethanol, and 70% ethanol, with the primary fatty acid identified as palmitic acid. The total flavonoid and polyphenol contents were 727.70-769.87 mg quercetin equivalent (QE)/ g and 309.24-348.09 mg gallic acid equivalent (GAE)/g, respectively, in extracts obtained using hot water, 30% ethanol, and 70% ethanol extracts. Measurements of antioxidant activity confirmed that extracts obtained using hot water, 30% ethanol, and 70% ethanol extracts increased the antioxidant effect in 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity, 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging activity, and ferric reducing antioxidant power (FRAP) in a concentration-dependent manner. A high correlation was observed between the total flavonoid content, total polyphenol content, and antioxidant activities of the extracts. This study provides data and novel insights for the development of functional food materials using P. yedoensis Matsum. bark extracts.

• Food Science and Preservation
• /
• v.31 no.4
• /
• pp.673-681
• /
• 2024

This study was conducted to develop kombucha with better functionality. The developed kombucha (CK) was prepared using the sugar extracts from fruits of Cudrania tricuspidata (Carrière) Bureau ex Lavallée instead of the sugar, which is used as a substrate for SCOBY in conventional kombucha (K). During fermentation, the soluble solids content significantly decreased in CK compared to K, and the pH change decreased rapidly in CK compared to K. On the 14^th day of fermentation, the weight of the SCOBY in CK was higher than that in K. Immediately after preparation, K contained only sucrose, but CK contained sucrose, glucose, and fructose. SCOBY appears to use glucose and fructose preferentially during fermentation. K contained acetic acid and citric acid right after preparation. However, as fermentation progressed, the composition changed to acetic acid, citric acid, and lactic acid. At the same time, CK initially consisted of citric acid, lactic acid, and acetic acid. However, acetic acid and citric acid increased but lactic acid decreased significantly on the 14^th day of fermentation. In the cytotoxicity studies, the CK showed a proliferation-promoting effect on normal lung cells (MRC-5) and strong cytotoxicity against human lung cancer cells (A549). These results suggest that the kombucha made from sugar extracts of C. tricuspidata fruits can be used as a more functional beverage than regular kombucha.

• Journal of Astronomy and Space Sciences
• /
• v.41 no.3
• /
• pp.171-194
• /
• 2024

This study reports comprehensive observations for the G5-level geomagnetic storm that occurred from May 10 to 12, 2024, the most intense event since the 2003 Halloween storm. The storm was triggered by a series of coronal mass ejections (CMEs) originating from the merging of two active regions 13664/13668, which formed a large and complex photospheric magnetic configuration and produced X-class flares in early May 2024. Among the events, the most significant CME, driven by an X2.2 flare on May 9, caught up with and merged with a preceding slower CME associated with an X-class flare on May 8. These combined CMEs reached 1 AU simultaneously, resulting in an extreme geomagnetic storm. Geostationary satellite observations revealed changes in Earth's magnetosphere due to solar wind impacts, increased fluxes of high-energy particles, and periodic magnetic field fluctuations accompanied by particle injections. Extreme geomagnetic storms resulting from the interaction of the solar wind with the Earth's magnetosphere caused significant energy influx into Earth's upper atmosphere over the polar regions, leading to thermospheric heating and changes in the global atmospheric composition and ionosphere. As part of this global disturbance, significant disruptions were also observed in the East Asian sector, including the Korean Peninsula. Ground-based observations show strong negative storm effects in the ionosphere, which are associated with thermospheric heating and resulting in decreases in the oxygen-to-nitrogen ratio (O/N₂) in high-latitude regions. Global responses of storm-time prompt penetration electric fields were also observed from magnetometers over the East-Asian longitudinal sector. We also briefly report storm-time responses of aurora and cosmic rays using all-sky cameras and neutron monitors operated by the Korea Astronomy and Space Science Institute (KASI). The extensive observations of the G5-level storm offer crucial insights into Sun-Earth interactions during extreme space weather events and may help establish better preparation for future space weather challenges.

• Journal of Korean Home Economics Education Association
• /
• v.35 no.3
• /
• pp.47-66
• /
• 2023

The purpose of this study was to develop an educational smart app for classes by reconstructing some of the teaching-learning contents of the clothing preparation planning within the 'clothing preparation planning and selection' curriculum unit. To this end, a teaching-learning process plan was planned for the classes, a smart app was developed for classes, and feedback from home economics teachers and app development experts was received for the developed app. The main composition of the developed app consists of five steps. The first step is to set up a profile using a real photo, ZEPETO or Galaxy emoji, or iPhone Memoji. In the second step, students make a list of clothes by figuring out the types, quantities and conditions of their exisitng wardrobe items. Each piece of clothing is assigned an individual registration number, and stduents can take pictures of the front and back, along with describing key attributes such as type, color, season-appropriateness, purchase date, and current status. Step three guides students in deciding which garments to retain and which to discard. Building on the clothing inventory from the previous step, students classify items to keep and items to dispose of. In Step 4, Deciding How to Arrange Clothing, students decide how to arrange clothing by filling out an alternative scorecard. Through this process, students can learn in advance the subsection of resource management and self-reliance, laying the foundationa for future learning in 'Practice of Rational Consumption Life'. Lastly, in the fifth stage of determining the disposal method, this stage is to develop practical problem-oriented classes on how to dispose of the clothes to be discarded in the thirrd stage by exploring various disposal methods, engaging in group discussions, and sharing opinions. This study is meaningful as a case study as an attempt to develop a smart app for education by an instructor to align teaching plans and educational content with achievement standards for the class. In the future, upgrades will have to be made through user application.

• Tunnel and Underground Space
• /
• v.34 no.4
• /
• pp.374-392
• /
• 2024

In order to successfully select a site for deep geological disposal of high-level radioactive waste, it is important to perform the stepwise approach along with the systematic selection and survey of evaluation parameters of geological environmental characteristics suitable for the domestic geological environment. In this study, we evaluated the characteristics of hydraulic conductivity, which is considered the most important evaluation parameter in the field of hydrogeology, targeting a deep-depth rock aquifer where actual disposal facilities are expected to be located. In particular, for the first time in Korea, we obtained in-situ pressure-flow data by directly conducting hydraulic tests in boreholes at depths ranging from 500 m to 750 m in various rock types distributed in Korea (granite/volcanic rock/gneiss/mudstone). And we derived hydraulic conductivity values by rock types and depth using verified analytical methods. For this purpose, precision hydraulic testing equipment developed in-house through this study was used, and detailed investigation procedures based on standard test methods were applied to field tests. As a result of the analysis, the average hydraulic conductivity value was found to be in the range of 10^-9 m/s in all granite/volcanic rock/gneiss areas. In the mudstone area, an average hydraulic conductivity value of 10^-11 m/s was derived, which was about 100 times (2 orders of magnitude) lower than that of the fractured rock aquifers. Moreover, permeability tended to slightly decrease with depth in fractured rock aquifers (granite and volcanic rock areas) containing many rock fractures. The gneiss area tended to have large local differences in permeability according to the composition of the stratum and the development of fracture zones rather than depth. In mudstone areas with weak fracture development, there was no significant variation in rock permeability according to depth. The hydraulic conductivity results by various rock types and depth presented in this study are expected to be utilized in building a foundational database for the site selection, design, and construction of disposal facilities in Korea.

• Journal of Space Technology and Applications
• /
• v.4 no.3
• /
• pp.210-219
• /
• 2024

This study investigates various ionospheric and thermospheric disturbances around the Korean Peninsula during the G5 geomagnetic storm occurred on May 10, 2024. This level of storm was the first of its magnitude in 21 years, resulting in auroras visible even in South Korea and severe space weather worldwide. The Korea Astronomy and Space Science Institute has been providing ionospheric information over Korea through total electron content (TEC) measurements from the Global Navigation Satellite System (GNSS) and monitoring the impact of ionospheric disturbances on GNSS signals by operating five GNSS scintillation stations in Korea and other countries. During this storm period, large amplitudes of TEC variations were observed over South Korea, along with anomalous TEC enhancements accompanied by strong scintillations at night and persistent TEC depletion on the dayside during the storm's recovery phase. Such daytime TEC depletion disturbances are quite rare, typically occurring only a few times throughout the 11-year solar cycle. While the association of persistent TEC depletion during the daytime with neutral composition disturbances was identified through observations, the causes of TEC enhancement and strong scintillation at night remain unclear. We speculate that the uplift of the ionosphere by storm-induced electric fields is responsible for the TEC enhancement and scintillation, but this hypothesis requires validation based on additional observational data.

• Korean Journal of Environment and Ecology
• /
• v.38 no.4
• /
• pp.375-387
• /
• 2024

This study investigated the ecological characteristics of Fraxinus chiisanensis Nakai, one of the endemic and rare plant species in Korea, based on its distribution status, characteristics of the growth environment, and species composition. A vegetation survey that analyzed the correlation between species distribution patterns and environmental variables, along with the traits of the emergent plant species, was performed according to the explanation of environmental growth conditions and phytosociological method for the location where F. chiisanensis is found. A total of 19 dominant locations and 9 non-dominant locations of F. chiisanensis were observed in 28 study sites in 12 regions, and a total of 155 taxa were observed. According to the vegetation climate of Korea, the growth environment of the study site where F. chiisanensis is located is characterized as cold and is primarily situated within the northern temperate deciduous broadleaf forest zone. The average elevation was 859m above sea level, with an average rock exposure of 60.4%, soil exposure of 24.7%, and an average slope of 18.7°. The taxa belonging to the top P-NCD(Percentage of Net Contribution Degree) among the emergent species were mostly designated as the taxa emerging in valley vegetation. The correlation analysis of environmental variables revealed that altitude had the strongest correlation, with rock exposure showing the second highest correlation. The ongoing dynamics of the F. chiisanensis forest are anticipated to persist due to the high P-NCD values exhibited by the F. chiisanensis within the shrub and herbaceous layers among the taxa associated with tree species. Most F. chiisanensis habitats are currently situated within protected regions such as national parks, provincial parks, and county parks, where there are relatively minimal human-induced disturbances. However, there is potential for damage in areas not designated as protected, such as forest tending operation sites or new hiking trails. Concerns about declining habitat quality have prompted suggestions for management strategies such as establishing Forest Genetic Resource Reserves in these locations. In addition, follow-up and further research should be conducted to identify possible sites for distribution and establish candidate conservation areas based on various environmental conditions of F. chiisanensis.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.42 no.1
• /
• pp.35-44
• /
• 2024

The purpose of this study is to investigate the garden architectural characteristics of the Villa Medici a Fiesole, which occupies an important place in the Italian Renaissance garden, and it is concluded as follows. First, the Villa Medici a Fiesole was built between 1451 and 1457 for Giovanni, the son of Cosimo, and then owned by many people and came to the present. It was designed by Michelozzo and is the first newly built Renaissance villa to show Alberti's concept of rural residence. Second, the location is the biggest characteristic of the Villa, which was built on the steep slope of the Fiesole and offers a panoramic view of the landscape of Florence. Third, the overall composition of the garden created using the slope is based on three terraces of different heights, and the overall spatial arrangement is made up of an orthogonal axis at which the vertical and horizontal axes intersect around the villa. Fourth, in the case of vegetation and objects, old paulownia, roses, and potted lemon trees are placed on the first terrace on the three lawn flower beds. Fountain and niche are located at the middle and east end of the terrace. The second terrace cannot be reached through stairs, but only through a villa or a ramp outside the door. The second terrace has a wisteria pergola that leads to a secret garden, and in the center of the pergola, there is a staircase that leads to the third terrace. The Secret Garden takes the form of a quarter garden and consists of an oval pond, a coniferous flowerbed, and magnolias. On the third terrace at the bottom, a symmetrically geometric flowerbed spreads out on both sides of a circular fountain, and large lemon trees are in the center of each flowerbed and four magnolias are planted around the fountain.