• Journal of The Korean Society of Agricultural Engineers
• /
• v.64 no.1
• /
• pp.15-26
• /
• 2022

3D open-field farm model developed from UAV (Unmanned Aerial Vehicle) data could make crop monitoring easier, also could be an important dataset for various fields like remote sensing or precision agriculture. It is essential to separate crops from the non-crop area because labeling in a manual way is extremely laborious and not appropriate for continuous monitoring. We, therefore, made a 3D open-field farm model based on UAV images and developed a crop segmentation model using a supervised machine learning algorithm. We compared performances from various models using different data features like color or geographic coordinates, and two supervised learning algorithms which are SVM (Support Vector Machine) and KNN (K-Nearest Neighbors). The best approach was trained with 2-dimensional data, ExGR (Excess of Green minus Excess of Red) and z coordinate value, using KNN algorithm, whose accuracy, precision, recall, F1 score was 97.85, 96.51, 88.54, 92.35% respectively. Also, we compared our model performance with similar previous work. Our approach showed slightly better accuracy, and it detected the actual crop better than the previous approach, while it also classified actual non-crop points (e.g. weeds) as crops.

• Smart Media Journal
• /
• v.5 no.3
• /
• pp.17-24
• /
• 2016

Korean agriculture is currently experiencing difficulties as a cause of rural depopulation, aging of rural population, grain self-sufficiency rate decline, and deepening of climate change. It is necessary to ensure our country's agriculture industrial competitiveness in accordance with opening of FTA imports expanded. To ensure the underdeveloped competitive, Korean government defines the 3rd generation model from 1st generation model to extend the smart farms of Korean types. The agriculture smarting overcomes the growth limitations of agriculture, and efforts to develop 6th + ${\alpha}$ industry. In this paper, We define and verify the IoF(Internet of Farming)-Cloud based substantial services about 2rd generation model, and propose a greenhouse of IoF-Cloud testbed.

• Research in Plant Disease
• /
• v.28 no.3
• /
• pp.152-161
• /
• 2022

Several types of chemical bactericides have been used to control fire blight. However, their excessive usage leads to environmental deterioration. Therefore, several researchers have analyzed antagonistic microorganisms as promising, effective, and safe biological control agents (BCAs). The primary aim of this study was to screen for potential antagonistic bacteria that suppress Erwinia amylovora. Among the 45 isolates studied, 5 strains showed the largest inhibition zone against E. amylovora. 16S rRNA gene sequencing identified them as Bacillus amyloliquefaciens (KPB 15), B. stratosphericus (KPB 21), B. altitudinis (KPB 25), B. safensis (KPB 31), and B. subtilis (KPB 39). KPB 25 and 31 reduced the lesion size of fire blight by 50% in immature apple fruits, and did not show antagonism against each other. Therefore, KPB 25 and 31 were selected to develop an antagonistic mixture against fire blight. Although the mixture with KPB 25 and 31 showed a slightly increased ability to reduce lesion size on immature fruits, they did not exhibit a synergistic effect in reducing E. amylovora population compared to each strain alone. Nevertheless, we have identified these two strains as useful and novel BCAs against fire blight with additional benefits safety and potential in developing a mixture without loss of their activity, owing to the absence of antagonism against each other.

• Journal of Drive and Control
• /
• v.20 no.3
• /
• pp.15-24
• /
• 2023

The aim of this study was to measure and evaluate the exhaust emission factors of agricultural tractors. Engine characteristics and three exhaust emissions (CO, NO_x, PM) were collected under actual agricultural operating conditions. Experiments were performed on idling, driving, plow tillage, and rotary tillage. The load factor (LF) was calculated using the collected engine data, and the emission factor was analyzed using the LF and exhaust emissions. The engine characteristics and exhaust emissions were significantly different for each working condition, and in particular, the LF was significantly different from the currently applied 0.48 LF. The data distribution of exhaust emissions was different depending on the engine speed. In some conditions, the emission factor was higher than the exhaust emission standards. However, since most emission limit standards are values calculated using an engine dynamometer, even if the emission factor measured under actual working conditions is higher, it cannot be regarded as wrong. It is expected that the results of this study can be used for the inventory construction of a calculation for domestic agricultural machinery emissions in the future.

• Korean Journal of Agricultural Science
• /
• v.50 no.4
• /
• pp.617-628
• /
• 2023

This study aimed to evaluate the engine load factor (LF) of a 90 kW agricultural combine harvester. The combine harvester used in this study is equipped with an electronic engine, and real-time engine data (torque and speed) was collected through a controller area network. The speed of the combine harvester during harvesting operation was divided into three levels (4, 5, and 6 km/h) for the representative operation speed range of 4 to 6 km/h. The LF was calculated using the engine load data measured in real time during harvesting. A weight was applied to the LF for each condition based on a survey of the usage. Results of the field test showed that the LF was 0.53, 0.64, and 0.87 at working speeds of 4, 5, and 6 km/h, respectively. The highest engine load factor was recorded at 6 km/h. Finally, based on the weight for the usage applied, the integrated engine LF was analyzed to be 0.69, which is approximately 144% higher than the currently applied LF of 0.48. A study on LF analysis for the entire work cycle, including idling and driving of the combine harvester, will be addressed in a future study.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.04a
• /
• pp.101-101
• /
• 2022

• Korean Journal of Organic Agriculture
• /
• v.31 no.3
• /
• pp.229-247
• /
• 2023

This study investigates the facility use status of smart farming farmers to improve facility use rate of farmers. To this end, a survey was conducted on smart farming farmers in Jeonnam province, and the main survey contents are as follows: facility use rate, the reasons for low facility use, the perception of the introduction and use of smart farming etc. As a result of the survey, many farmers have introduced smart farming facilities even though they do not have enough use capacity. Thus it is necessary to improve the use capacity of farmers. Second, the average facility use rate of farmers was 65.1%, and 37.5% of respondents did not use even 50% of smart farming facilities. To improve the use rate, education on how to use facilities and continuous consulting support for farmers are needed. And the largest number of farmers perceived the risk like crop damage or facility failure due to poor use of facilities. This means that risk management due to the smart farming facilities is important. Third, farmers answered that rapid and continuous repair service were the most important when using facilities. Thus it is important to foster rear industries such as maintenance companies to stably operate smart farming facilities.

• Research in Plant Disease
• /
• v.26 no.3
• /
• pp.121-133
• /
• 2020

Climate change, increased extreme weather and climate events, and rapidly changing socio-economic environment threaten agriculture and thus food security of our society. Therefore, it is urgent to shift from conventional farming to smart agriculture using big data and artificial intelligence to secure sustainable growth. In order to efficiently manage plant diseases through smart agriculture, agricultural big data that can be utilized with various advanced technologies must be secured first. In this review, we will first learn about agrometeorological big data consisted of meteorological, environmental, and agricultural data that the plant pathology communities can contribute for smart plant disease management. We will then present each sequential components of the smart plant disease management, which are prediction, monitoring and diagnosis, control, prevention and risk management of plant diseases. This review will give us an appraisal of where we are at the moment, what has been prepared so far, what is lacking, and how to move forward for the preparation of smart plant disease management.

• Smart Media Journal
• /
• v.11 no.4
• /
• pp.9-18
• /
• 2022

Recently, research and system using AI is rapidly increasing in various fields. Smart farm using artificial intelligence and information communication technology is also being studied in agriculture. In addition, data-based precision agriculture is being commercialized by convergence various advanced technology such as autonomous driving, satellites, and big data. In Korea, the number of commercialization cases of facility agriculture among smart agriculture is increasing. However, research and investment are being biased in the field of facility agriculture. The gap between research and investment in facility agriculture and open-air agriculture continues to increase. The fields of fruit trees and plant factories have low research and investment. There is a problem that the big data collection and utilization system is insufficient. In this paper, we are proposed the system for determining the fruit tree growth stage using a deep learning-based object detection model. The system was proposed as a hybrid app for use in agricultural sites. In addition, we are implemented an object detection function for the fruit tree growth stage determine.

• Korean Journal of Agricultural Science
• /
• v.47 no.3
• /
• pp.615-624
• /
• 2020

An effective sampling method is necessary to monitor potato tuber moths (Phthorimaea operculella) because they are the biggest concern in potato-cultivating areas. In this study, a sequential sampling method was developed based on the results of field surveys of potato tuber moths in South Korea. Potato tuber moths were collected in fields cultivating potatoes at six sites, and their spatial distribution was investigated using the Taylor power law. The optimal sampling size and cumulative number of potato tuber moths in traps to stop sampling were determined based on the spatial distribution pattern and mean density of the collected potato tuber moths. Finally, the developed sampling method was applied to propose a control action, and its sampling efficiency was compared with that of the traditional sampling method using a binomial distribution. The potato tuber moths tended to aggregate; the optimal number was approximately 5 - 16 traps for sampling, and the number varied with the mean density of potato tuber moths according to the sampling sites. In addition, one, two, and three sites might require the following actions: Continued sampling, control, and no control, respectively. Sampling with the binomial distribution showed the minimum sample size was 12 when considering the economic threshold level. Here, we propose an effective sampling method that can be applied for future monitoring and field surveys of potato tuber moths in South Korea.