• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.3
• /
• pp.141-148
• /
• 2021

Agrivoltaic systems, also called solar sharing, stated from an idea that utilizes sunlight above the light saturation point of crops for power generation using solar panels. It is expected that agrivoltaic systems can realize climate smart agriculture by reducing evapotranspiration and methane emission due to the reduction of incident solar radiation and the consequent surface cooling effect and bring additional income to farms through solar power generation. In this study, to evaluate that agrivoltaic systems are suitable for realization of climate smart agriculture, we conducted agro-environmental observations (i.e., downward/upward shortwave/longwave radiations, air temperature, relative humidity, water temperature, soil temperature, and wind speed) in a rice paddy under an agrivoltaic system and compared with the environment outside the system using automated meteorological observing systems (AMOS). During the observation period, the spatially averaged incoming solar radiation under the agrivoltaic system was about 70% of that in the open paddy field, and clear differences in the soil and water temperatures between the paddy field under the agrivoltaic system and the open paddy field were confirmed, although the air temperatures were similar. It is required in the near future to confirm whether such environmental differences lead to a reduction in water consumption and greenhouse gas emissions by flux measurements.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.241-265
• /
• 2023

A corporate insolvency prediction model serves as a vital tool for objectively monitoring the financial condition of companies. It enables timely warnings, facilitates responsive actions, and supports the formulation of effective management strategies to mitigate bankruptcy risks and enhance performance. Investors and financial institutions utilize default prediction models to minimize financial losses. As the interest in utilizing artificial intelligence (AI) technology for corporate insolvency prediction grows, extensive research has been conducted in this domain. However, there is an increasing demand for explainable AI models in corporate insolvency prediction, emphasizing interpretability and reliability. The SHAP (SHapley Additive exPlanations) technique has gained significant popularity and has demonstrated strong performance in various applications. Nonetheless, it has limitations such as computational cost, processing time, and scalability concerns based on the number of variables. This study introduces a novel approach to variable selection that reduces the number of variables by averaging SHAP values from bootstrapped data subsets instead of using the entire dataset. This technique aims to improve computational efficiency while maintaining excellent predictive performance. To obtain classification results, we aim to train random forest, XGBoost, and C5.0 models using carefully selected variables with high interpretability. The classification accuracy of the ensemble model, generated through soft voting as the goal of high-performance model design, is compared with the individual models. The study leverages data from 1,698 Korean light industrial companies and employs bootstrapping to create distinct data groups. Logistic Regression is employed to calculate SHAP values for each data group, and their averages are computed to derive the final SHAP values. The proposed model enhances interpretability and aims to achieve superior predictive performance.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.3
• /
• pp.226-235
• /
• 2023

This study aimed to supplement the shortcomings of the Multiple-sensor-based Frost Observation System (MFOS). The developed frost observation system is an improvement of the existing system. Based on the leaf wetness sensor (LWS), it not only detects frost but also functions to predict surface temperature, which is a major factor in frost occurrence. With the existing observation system, 1) it is difficult to observe ice (frost) formation on the surface when capturing an image of the LWS with an RGB camera because the surface of the sensor reflects most visible light, 2) images captured using the RGB camera before and after sunrise are dark, and 3) the thermal infrared camera only shows the relative high and low temperature. To identify the ice (frost) generated on the surface of the LWS, a LWS that was painted black and three sheets of glass at the same height to be used as an auxiliary tool to check the occurrence of ice (frost) were installed. For RGB camera shooting before and after sunrise, synchronous LED lighting was installed so the power turns on/off according to the camera shooting time. The existing thermal infrared camera, which could only assess the relative temperature (high or low), was improved to extract the temperature value per pixel, and a comparison with the surface temperature sensor installed by the National Institute of Meteorological Sciences (NIMS) was performed to verify its accuracy. As a result of installing and operating the MFOS v2, which reflects these improvements, the accuracy and efficiency of automatic frost observation were demonstrated to be improved, and the usefulness of the data as input data for the frost prediction model was enhanced.

• Korean Journal of Environment and Ecology
• /
• v.25 no.6
• /
• pp.893-902
• /
• 2011

The purpose of this study was to survey structures and growth conditions of Pinus densiflora village groves, and to establish management strategies for their desirable growth and conservation. Twelve village groves were selected in western Gangwon province for the study. The age of the study groves ranged from 50 to 200 years. Average dbh (diameter at breast height) and density of trees for each study grove were 27~52cm and 0.5~9.3 trees/$100m^2$, respectively. Soil environments were favorable to Pinus densiflora growth in the majority of the study groves, but 2 study groves with sandy soils showed considerably poor nutrient contents. Low tree vitality was found in some of the study groves due to poor conditions of root growth from soil fill and trampling. There were detachment of cambial tissue and damage of stem cavity at 6 study groves, which were caused by artificial injury, careless pruning, and frost damage. Light disease damage by Rhizosphaera kalkhoffii and phomopsis blight were found at 6 study groves. Light pest damage by Thecodiplosis japonensis was also found at 6 study groves, but the pest damage at 2 study groves was relatively considerable. Thus, major factors limiting normal growth of Pinus densiflora village groves were infertility, soil fill and trampling, stem damage, and disease and pest. Desirable management strategies were explored to solve growth-related problems and to conserve the study groves. The management strategies included fertilization of organic matter and lime, removal of soil fill, soil plowing and graveling, wood-trail installation or woodchip mulching, supply of wood fences and protective frames, surgical operation for damaged stems, vitality enhancement, and trunk injection to improve growth environments or control stem damage and disease/pest.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.5
• /
• pp.1-6
• /
• 2020

UAV (Unmanned Aerial Vehicles) are used widely for military purposes because they are more economical than general manned aircraft and satellites, and have easy access to the object. Recently, owing to the development of IT technology, UAV equipped with various sensors have been released, and their use is increasing in a wide range of fields, such as surveying, agriculture, meteorological observation, communication, broadcasting, and sports. An increasing number of studies and attempts have made use of it. On the other hand, existing research was related mostly to photogrammetry, but there has been a lack of analytical research on LiDAR (Light Detection And Ranging). Therefore, this study examined the characteristics of a UAV LiDAR sensor for the application of a geospatial information field. In this study, the performance of commercialized LiDAR sensors, such as the acquisition speed and the number of echoes, was investigated, and data acquisition and analysis were conducted by selecting Surveyor Ultra and VX15 models with similar accuracy and data acquisition distances. As a result, a DSM of each study site was generated for each sensor, and the characteristics of data density, precision, and acquisition of ground data from vegetation areas were presented through comparison. In addition, the UAV LiDAR sensor showed an accuracy of 0.03m ~ 0.05m. Hence, it is necessary to select equipment considering the characteristics of data for effective use. In the future, the use of UAV LiDAR may be suggested if additional data can be obtained and analyzed for various areas, such as urban areas and forest areas.

• Korean Journal of Soil Science and Fertilizer
• /
• v.35 no.3
• /
• pp.185-195
• /
• 2002

The purpose of this study is to evaluate and seek the best suitable band and threshold boundary level on the change detection of image differencing method using Landsat TM data(20 May 1987 and 20 May 1993) in Pyeongtaeg City. The change detection images differencing method were evaluated by using normal reference data with an optimal threshold level{$mean{\pm}(SD{\times}T$ value). The normal reference data consisted of positive change{change dark into light in image pattern, that is, it changed arable land(paddy, upland, forest and so on) to artificial area(buildings, vinyl-house and roads, etc)} and negative change(change light into dark in image pattern, that is, it changed artificial area into arable land). As the result, the kappa coefficients of visible bands(D1, D2 and D3) were higher than those of infrared bands(D4, D5 and D7), and than D1 image with 1.0 thresholding and normal reference data was a improved result in the land-surface change detection such as kappa coefficient : 68.4%, overall accuracy : 89.2%, negative change : 6.6%, positive change : 10.6%.

• Safety and Health at Work
• /
• v.10 no.4
• /
• pp.452-460
• /
• 2019

Background: Monitoring and control of PM_2.5 are being recognized as key to address health issues attributed to PM_2.5. Availability of low-cost PM_2.5 sensors made it possible to introduce a number of portable PM_2.5 monitors based on light scattering to the consumer market at an affordable price. Accuracy of light scatteringe-based PM_2.5 monitors significantly depends on the method of calibration. Static calibration curve is used as the most popular calibration method for low-cost PM_2.5 sensors particularly because of ease of application. Drawback in this approach is, however, the lack of accuracy. Methods: This study discussed the calibration of a low-cost PM_2.5-monitoring device (PMD) to improve the accuracy and reliability for practical use. The proposed method is based on construction of the PM_2.5 sensor network using Message Queuing Telemetry Transport (MQTT) protocol and web query of reference measurement data available at government-authorized PM monitoring station (GAMS) in the republic of Korea. Four machine learning (ML) algorithms such as support vector machine, k-nearest neighbors, random forest, and extreme gradient boosting were used as regression models to calibrate the PMD measurements of PM_2.5. Performance of each ML algorithm was evaluated using stratified K-fold cross-validation, and a linear regression model was used as a reference. Results: Based on the performance of ML algorithms used, regression of the output of the PMD to PM_2.5 concentrations data available from the GAMS through web query was effective. The extreme gradient boosting algorithm showed the best performance with a mean coefficient of determination (R²) of 0.78 and standard error of 5.0 ㎍/㎥, corresponding to 8% increase in R² and 12% decrease in root mean square error in comparison with the linear regression model. Minimum 100 hours of calibration period was found required to calibrate the PMD to its full capacity. Calibration method proposed poses a limitation on the location of the PMD being in the vicinity of the GAMS. As the number of the PMD participating in the sensor network increases, however, calibrated PMDs can be used as reference devices to nearby PMDs that require calibration, forming a calibration chain through MQTT protocol. Conclusions: Calibration of a low-cost PMD, which is based on construction of PM_2.5 sensor network using MQTT protocol and web query of reference measurement data available at a GAMS, significantly improves the accuracy and reliability of a PMD, thereby making practical use of the low-cost PMD possible.

• Proceedings of the Plant Resources Society of Korea Conference
• /
• 2018.04a
• /
• pp.22-22
• /
• 2018

실고사리[Lygodium japonicum (Thunb.) Sw.]는 국내 자생하는 덩굴성 양치식물로 유인재배가 용이하여 실내 외 조경 및 관상소재로 활용이 가능하다. 한방에서는 전초를 해금사초, 포자를 해금사, 뿌리와 지하경을 해금사근이라 약재로 사용한다. 본 연구는 관상 및 약재로 이용이 가능한 실고사리의 대량생산을 위한 기내 외 번식방법을 개발하고자 수행되었다. 식물재료는 경상북도 의성군 일대에서 성체를 수집하여 청주의 일반하우스에 식재하여 성숙한 포자를 채취하였다. 포자를 기내 발아시켜 전엽체를 획득한 다음 8주 간격으로 계대하면서 연구의 재료를 확보하였다. 전엽체의 증식과 생육에 적합한 배지를 비교 하고자, 1/4, 1/2, 1, 2MS배지와 Knop배지를 조성하여 배양하였다. 배양방법은 전엽체 300mg을 메스로 균일하게 다져서 배양하는 방법을 이용하였으며, 배양환경은 온도($25{\pm}1.0^{\circ}C$), 광도($30{\pm}1.0{\mu}mol{\cdot}m^{-2}{\cdot}s^{-1}$), 광주기(16/8h light/dark)로 조절되었다. 연구결과, 모든 처리구의 생체중은 초기접종량 보다 11배 이상 증가하였고, 그 중 1MS배지에서는 전엽체의 생체중이 7.3g으로 가장 많이 증가하였다. 뿐만 아니라, 형태형성도 우수하여 모두 정상적인 전엽체의 형태인 하트형으로 발달이 유도되었다. 전엽체로부터 포자체 형성을 유도하고자, 원예상토, 피트모스, 펄라이트 및 마사토의 혼합비율을 5종류로 달리하여 사각분($7.5{\times}7.5{\times}7.5cm$)에 혼합토양을 충진하였다. 전엽체 1g과 증류수 25mL를 핸드블랜더로 10초간 분쇄하여 사각분의 토양표면에 균일하게 분주하는 방법을 사용하였다. 이후 온도($25{\pm}1.0^{\circ}C$), 광도($43{\pm}2.0{\mu}mol{\cdot}m^{-2}{\cdot}s^{-1}$), 광주기(16/8h light/dark), 습도($72{\pm}2.0%$)를 유지하면서 10주간 재배되었다. 연구결과, 원예상토 단용, 원예상토와 펄라이트 및 마사토가 2:1(v:v)로 혼합된 토양에서 각 498.0, 402.5, 482.5개의 포자체가 형성되어 사각분 면적대비 7.16개($cm^2$)가 생산되었다. 한편 포자체의 생육은 원예상토와 펄라이트가 2:1(v:v)로 혼합된 토양에서 생체중, 엽장, 엽폭, 근장 및 SPAD value 등이 우수하였다.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.45 no.2
• /
• pp.68-75
• /
• 2017

Demand for dwarf mondo grass (DMG; Ophiopogon japonicus 'Nanus') as an ornamental garden plant is expected to grow in the future. The purpose of this study was to investigate the levels of shade tolerance and ground cover by growing DMG under a variety of shade conditions for 18 months (May 2015~October 2016). DMG plants grown in bare ground for 3 years in Jangheung-gun, Jeonnam were used for testing. In an experimental site created in Naju city in Jeonnam, the DMG was planted in planters ($70cm{\times}70cm{\times}24cm$) and covered with a shading curtain to block natural light. Shaded conditions were then arranged under different levels of shade (0%, 55% and 75%). When the plants were grown, growth (leaf size, the number of leaves, fresh weight and dry weight) and ground coverage of DMG were analyzed. According to the results, DMG growth in terms of leaf size and the number of leaves was statistically higher under zero shade (full sunlight), when compared to other shaded conditions. DMG's fresh and dry weights were significantly greater under 0% and 55% shade, compared to those under 75% shade. The degrees of shade tolerance required for normal growth of DMG were found in the range of 0~50%, meaning that more than 50% shade may decrease plant growth. There were no statistical differences in ground coverage rates of DMG under different levels of shade. When 220 tillers were planted per $1m^2$ of plot, up to 80% of the area was covered by DMG after 18 months. Since DMG requires nutrient-rich soil to grow, sufficient nitrogen fertilizers are proposed to accelerate the ground cover of DMG. As DMG remained alive over the winter in the experiments, this study also suggests that DMG can be planted in the southern temperate region.

• Journal of Plant Biotechnology
• /
• v.49 no.3
• /
• pp.207-212
• /
• 2022

The optimal culture conditions for root organogenesis from the callus of peonies (Paeonia lactiflora Pall.) were investigated. Root explants with vascular bundles were cultured in Murashige and Skoog (MS) medium combined with 0.5-4.0 mg/L auxins (indole acetic acid [IAA], naphthalene acetic acid [NAA], indolebutyric acid [IBA], and 2,4-dichlorophenoxyacetic acid [2,4-D]) and 0.0-2.0 mg/L cytokinins (kinetin, zeatin, and benzylaminopurine [BAP]) to induce callus formation. The callus was then cultured in MS medium combined with three concentrations (0.1, 0.5, and 1.0 mg/L) of IAA, NAA, IBA, kinetin, zeatin, and BAP in the dark for 6 weeks. Based on the results, the effects of dark and light conditions on the callus cultured in MS medium with combinations of 0.1-1.0 mg/L IBA and zeatin for 6 weeks were studied. Callus formation was most effective (>+++) in the medium with a combination of 1.0 mg/L NAA and 1.0 mg/L zeatin. A high number of long adventitious roots were observed in the mediums with 0.1 mg/L IBA (6.66 and 4.82 cm) and 0.5 mg/L zeatin (2.32 and 0.72 cm) among auxins and cytokinins, respectively. The highest number (14.06) of adventitious roots were formed from the callus cultured in light in the MS medium combined with 0.1 mg/L IBA and 0.5 mg/L zeatin. This same medium induced the formation of the longest adventitious root (5.45 cm) in the dark. Thus, optimization of in vitro culture conditions may be possible for the mass propagation of adventitious roots in peonies.