• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.67-77
• /
• 2023

This study represents an innovative research conducted in the smart farm environment, developing a deep learning-based disease and pest detection model and applying it to the Intelligent Internet of Things (IoT) platform to explore new possibilities in the implementation of digital agricultural environments. The core of the research was the integration of the latest ImageNet models such as Pseudo-Labeling, RegNet, EfficientNet, and preprocessing methods to detect various diseases and pests in complex agricultural environments with high accuracy. To this end, ensemble learning techniques were applied to maximize the accuracy and stability of the model, and the model was evaluated using various performance indicators such as mean Average Precision (mAP), precision, recall, accuracy, and box loss. Additionally, the SHAP framework was utilized to gain a deeper understanding of the model's prediction criteria, making the decision-making process more transparent. This analysis provided significant insights into how the model considers various variables to detect diseases and pests.

• Journal of Industrial Convergence
• /
• v.22 no.2
• /
• pp.23-32
• /
• 2024

This study utilized minimum number of demographic variables and first-semester GPA of students to predict the final academic status of students at a vocational college in Seoul. The results from XGBoost and LightGBM models revealed that these variables significantly impacted the prediction of students' dismissal. This suggests that early academic performance could be an important indicator of potential academic dropout. Additionally, the possibility that academic years required to award an associate degree at the vocational college could influence the final academic status was confirmed, indicating that the duration of study is a crucial factor in students' decisions to discontinue their studies. The study attempted to model without relying on psychological, social, or economic factors, focusing solely on academic achievement. This is expected to aid in the development of an early warning system for preventing academic dropout in the future.

• Journal of agricultural medicine and community health
• /
• v.43 no.1
• /
• pp.9-17
• /
• 2018

Objectives: This study was to assess the status and awareness of excessive heat exposure among agricultural workers. Methods: We selected a total of 90 farmers from a villages of Gyeongju-si, during August, 2015. We carried out the temperature measurement for nine times and derived Health Index (HI) and Wet Bulb Globe Temperature (WBGT) index. We compared the HI, WBGT and excessive heat warnings. Status of high temperature exposure, lifestyle, medical history, and awareness about excessive health related exposure illness assessed using survey questionnaires. Results: The matching rates between the WBGT and the HI during excessive heat warning were high, but when it was a non-excessive heat warning, there were days of excessive HI or WBGT. Out of 90 farmers surveyed, 78 cases (86.7%) were in their 60s and older age group. Slightly more than two third (71.1%) farmers were farming in the dawn-morning (71.1%), and the daily working hours were less than 4 hours (54.4 %), but only 23.3% among farmers took regular breaks. Of total, 14.4% farmers experienced excessive heat exposure related illness in order of tiredness, lethargy, dizziness, headaches, and sweating. Overall, the awareness of the danger for excessive heat and the heat wave warnings were high at 70.0% and 74.4%, respectively. Conclusions: Politically, the excessive heat warnings should not be taken into account the simple temperature measurement but, have to consider WBGT and HI standards at the same time. Farmers need to be promoted and educated to prevent the excessive heat related illness by periodically increasing their rest time during farming.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.05b
• /
• pp.806-810
• /
• 2005

홍수로 인한 국민의 재산과 인명피해를 방지 또는 최소화하기 위한 방법으로 주요지점을 선정하여 홍수예보를 수행하고 있다. 이러한 홍수예보는 초기 강우법, 수위법 등의 단순한 방법으로 수행되었으나, 컴퓨터가 발달되면서 여러 형태의 홍수유출모형이 개발되어 현재에는 홍수유출모형을 이용한 홍수예보를 수행하고 있다. 낙동강홍수예보 시스템은 1987년 낙동강홍수통제소의 개소시 42개소의 수위관측소와 54개소의 우량관측소를 중심으로 43개 소유역을 구분하고 티센망을 구성하여 홍수예보시스템을 구축하였다. 홍수예보시스템이 구축된 후 여러 차례의 매개변수 개선과정을 거쳐 운용하고 있다. 반면, 지속적인 수문관측소의 증설이 이루어졌으나 이를 모형에 반영하지 않아 증설된 수위관측소를 반영하여 114개 소유역으로 구분하고 추가된 우량관측소를 이용하여 티센망을 구성하였으며, 당초 반영되지 않았던 소규모댐을 모형에 추가하여 예측의 정확도를 재고하였다. 증설된 수문관측소를 모형에 반영과 함께 우량자료의 신뢰성 및 안정성을 확보하기 위한 우량관측소의 확장방안을 제시하였다. 본 연구에서 수행한 낙동강홍수예보 시스템의 소유역을 114개로 나누고 기존의 유효우량산정법, 유역유출모형 하도유출모형의 매개변수를 전통적인 방법으로 재산정하여 홍수기에 적용한 결과 당초의 홍수예보시스템으로 예측한 결과보다 소유역의 적절하게 재구성하는 것으로도 예측의 정밀도가 개선되었음을 확인할 수 있었다. 그러나, 소유역의 수와 하도의 수가 증가하면서 발생하는 오차를 감소시킬 필요성이 있으며, 이는 지속적인 시스템의 운용을 통하여 모형구성을 개선함으로서 어느 정도 개선될 것으로 판단된다. 그밖에 기존의 매개변수산정을 전통적인 방법에서 우리나라 하천 또는 낙동강유역의 하천에 적당한 방법을 개발하여 개선시킬 필요성이 있다.>$4.3\%$로 가장 근접한 결과를 나타내었으며, 총 유출량에서도 각각 $7.8\%,\;13.2\%$의 오차율을 가지는 것으로 분석되어 타 모형에 비해 실유량과의 차가 가장 적은 것으로 모의되었다. 향후 도시유출을 모의하는 데 가장 근사한 유출량을 산정할 수 있는 근거가 될 것이며, 도시재해 저감대책을 수립하는데 기여할 수 있을 것이라 판단된다.로 판단되는 대안들을 제시하는 예비타당성(Prefeasibility) 계획을 수립하였다. 이렇게 제시된 계획은 향후 과학적인 분석(세부평가방법)을 통해 대안을 평가하고 구체적인 타당성(feasibility) 계획을 수립하는데 토대가 될 것이다.{0.11R(mm)}(r^2=0.69)$로 나타났다. 이는 토양의 투수특성에 따라 강우량 증가에 비례하여 점증하는 침투수와 구분되는 현상이었다. 경사와 토양이 같은 조건에서 나지의 경우 역시 $Ro_{B10}(mm)=20.3e^{0.08R(mm)(r^2=0.84)$로 지수적으로 증가하는 경향을 나타내었다. 유거수량은 토성별로 양토를 1.0으로 기준할 때 사양토가 0.86으로 가장 작았고, 식양토 1.09, 식토 1.15로 평가되어 침투수에 비해 토성별 차이가 크게 나타났다. 이는 토성이 세립질일 수록 유거수의 저항이 작기 때문으로 생각된다. 경사에 따라서는 경사도가 증가할수록 증가하였으며 $10\% 경사일 때를 기준으로 $Ro(mm)=Ro_{10}{\times}0.797{\times}e^{-0.021s(\%)}$로 나타났다.천성 승모판 폐쇄 부전등을 초래하는 심각한 선천성 심질환이다. 그러나 진단 즉시 직접 좌관상동맥-대동맥 이식술로 수술적 교정을 해줌으로써 좋은 성적을 기대할 수 있음을 보

• Journal of Korea Water Resources Association
• /
• v.44 no.11
• /
• pp.875-887
• /
• 2011

The objective of this study is to develop and evaluate a Korean threshold runoff computation method. The selected study area is the Han-River basin and the stream channels in the study area are divided into 3 parts; natural channel and artificial manmade channel for small mountainous catchments, and main channel for master stream. The threshold runoff criteria for small streams is decided to 0.5 m water level increase from the channel bottom, which is the level that mountain climbers and campers successfully escape from natural flood damage. Threshold runoff values in natural channel of small mountainous area are computed by the results from the regional regression analysis between parameters of basin and stream channel, while those in artificial channel of small mountainous area are obtained from the data of basin and channel characteristics parameter. On the other hand, the threshold runoff values for master channel are used the warning flood level that is useful information for escaping guideline for riverside users. For verification of the threshold runoff computation method proposed in this study, three flash flood cases are selected and compared with observed values, which is obtained from SCS effective rainfall computation. The 1, 3, 6-hour effective rainfall values are greater than the corresponding threshold runoff values represents that the proposed computation results are reasonable.

• Journal of the Korean Institute of Gas
• /
• v.11 no.4
• /
• pp.41-46
• /
• 2007

The residential fuel cell system was modeled as a box-shaped chamber with vent openings, filled with various components such as reformer, desulfurizer, fuel cell stack and humidifier. When the vent openings are 1% of the total surface and hydrogen leakage 1%, hydrogen concentration is around 0.1% higher than the other regions from leak points in the chamber at 30 seconds and hydrogen concentration is increased from 0.3% to 0.7% in the upper region of the system after 200 seconds. When the vent openings are 1% of the total surface and hydrogen leakage 1%, 3%, 5%, the steady state result of CFD, 5% of hydrogen leakage is reached the lowest ignition limit in the system. When the vent openings are 2% of the total surface and hydrogen leakage 1%, hydrogen concentration is increased in the bottom of the system for 60 seconds. After 250 seconds, hydrogen concentration is reached the steady state in the system. As the vent opening of the total surface increased from 1% to 2%, averaged hydrogen mole fraction is under 1% in the system, however, upper regions of the system from the hydrogen leakage points are shown over 1% of hydrogen mole fraction.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.25 no.1
• /
• pp.1-8
• /
• 2019

In December 2017, many collision accidents of small vessels, such as those between oil refineries and fishing boats, occurred near Yeonghung-do in Incheon. In order to prevent marine casualties from small vessels, the government is striving to improve the safety capabilities of ship operators by strengthening education and improving the working environment. They are providing education and refining training regulations for fishermen operating vessels under 5 tons. However, the situation includes certain vulnerabilities. In this study, we propose a collision prevention algorithm for small vessels using the Wireless Access in Vehicular Environments (WAVE) communication system, which is a new communication technique to prevent collisions with small ships. The collision avoidance algorithm used is based on DCPA/TCPA. Research analyses, simulation experiments and questionnaires have been conducted to define the criteria of DCPA/TCPA. As a result, the standard for DCPA was $8(L_a+L_b)$ and for TCPA was 2.5 min. Three different accident cases were selected, and this algorithm was applied to confirm alarm responses at certain times. This algorithm can provide information to the operators of small ships in advance to help them recognize potential collision situations.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.3
• /
• pp.153-163
• /
• 2017

The objective of this study is to enhance the model's speed of estimating weather variables (e.g., minimum/maximum temperature, sunshine hour, PRISM (Parameter-elevation Regression on Independent Slopes Model) based precipitation), which are applied to the Agrometeorological Early Warning System (http://www.agmet.kr). The current process of weather estimation is operated on high-performance multi-core CPUs that have 8 physical cores and 16 logical threads. Nonetheless, the server is not even dedicated to the handling of a single county, indicating that very high overhead is involved in calculating the 10 counties of the Seomjin River Basin. In order to reduce such overhead, several cache and parallelization techniques were used to measure the performance and to check the applicability. Results are as follows: (1) for simple calculations such as Growing Degree Days accumulation, the time required for Input and Output (I/O) is significantly greater than that for calculation, suggesting the need of a technique which reduces disk I/O bottlenecks; (2) when there are many I/O, it is advantageous to distribute them on several servers. However, each server must have a cache for input data so that it does not compete for the same resource; and (3) GPU-based parallel processing method is most suitable for models such as PRISM with large computation loads.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.3
• /
• pp.149-155
• /
• 2021

Early warning systems for weather risk management in the agricultural sector have been developed to predict potential wind damage to crops. These systems take into account the daily maximum wind speed to determine the critical wind speed that causes fruit drops and provide the weather risk information to farmers. In an effort to increase the accuracy of wind risk predictions, an artificial neural network for binary classification was implemented. In the present study, the daily wind speed and other weather data, which were measured at weather stations at sites of interest in Jeollabuk-do and Jeollanam-do as well as Gyeongsangbuk- do and part of Gyeongsangnam- do provinces in 2019, were used for training the neural network. These weather stations include 210 synoptic and automated weather stations operated by the Korean Meteorological Administration (KMA). The wind speed data collected at the same locations between January 1 and December 12, 2020 were used to validate the neural network model. The data collected from December 13, 2020 to February 18, 2021 were used to evaluate the wind risk prediction performance before and after the use of the artificial neural network. The critical wind speed of damage risk was determined to be 11 m/s, which is the wind speed reported to cause fruit drops and damages. Furthermore, the maximum wind speeds were expressed using Weibull distribution probability density function for warning of wind damage. It was found that the accuracy of wind damage risk prediction was improved from 65.36% to 93.62% after re-classification using the artificial neural network. Nevertheless, the error rate also increased from 13.46% to 37.64%, as well. It is likely that the machine learning approach used in the present study would benefit case studies where no prediction by risk warning systems becomes a relatively serious issue.

• Korean Journal of Environmental Agriculture
• /
• v.31 no.2
• /
• pp.122-128
• /
• 2012

BACKGROUND: Mine drainage from metal mining districts is a well-recognized source of environmental contamination. Oxidation of metal sulfides in mines, mine dumps and tailing impoundments produces acidic, metal-rich waters that can contaminate the local surface water and soil. METHODS AND RESULTS: This experiment was carried out to investigate the pollution assessment of heavy metal on the water quality of mine drainage, paddy soils and sediment in lower watershed affected by mine drainage of the Sambo mine. The average concentrations of dissolved Cd (0.018~0.035 mg/L) in mine drainage discharged from the main waste rock dumps(WRD) was higher than the water quality standards (0.01 mg/L) for agricultural water in Korea. Also, the average concentrations of dissolved Zn, Fe and Mn were higher than those of recommended maximum concentrations (Zn 2.0, Fe 5.0, Mn 0.2 mg/L) of trace metal in irrigation water proposed by FAO (1994). The average contents of Pb and Zn in paddy soils was higher than those of standard level for soil contamination(Pb 200, Zn 300 mg/kg) in agricultural soil by Soil Environmental Conservation Act in Korea. Also, the concentrations of Cd, Pb and Zn in sediment were higher than those of standard level for soil contamination (Cd 10, Pb 400, Zn 600 mg/L) in waterway soil by Soil Environmental Conservation Act in Korea. The enrichment factor (EFc) of heavy metals in stream sediments were in the order as Cd>Pb>Zn> As>Cu>Cr>Ni. Also, the geoaccumulation index (Igeo) of heavy metals in stream sediments were in the order as Zn>Cd>Pb>Cu>As>Cr>Ni, specially, the geoaccumulation index (Igeo) of Zn (Igeo 3.1~6.2) were relatively higher than that of other metals in sediment. CONCLUSION(s): The results indicate that stream water and sediment were affected by mine drainage discharged from the Sambo mine at least to a distance of 1 km downstream (SN-1, SN-2) of the mine water discharge point.