• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.69 no.3
• /
• pp.154-162
• /
• 2024

In response to the increasing impact of climate change on agriculture, various cultivation technologies have been recently developed to improve agricultural productivity and reduce carbon emissions for carbon neutrality. This study presents an algorithm for estimating rice planting density in agriculture using drone-captured images and deep learning-based image analysis technology. The algorithm utilizes images collected from various paddies; these images are processed through pre-processing steps and serve as training data for the YOLOv5x deep learning model. The trained model demonstrated high precision and recall, effectively estimating the position information of rice plants in each image. By accurately estimating the position of rice plants based on the central coordinates in diverse unpaved environments, the model allowed for estimation of rice plant density in each paddy, producing values closely aligned with actual measurements. Moreover, the algorithm proposed in this study provides a novel approach for precise determination of rice planting density based on the position information of rice plants in the images. Analysis of drone footage from different regions capturing portions of paddies revealed that the developed algorithm exhibited a significant correlation (R² =0.877) with actual planting density. This finding suggests the potential effective application of the algorithm in real-world agricultural settings. In conclusion, we believe that this research contributes to the ongoing digital transformation in agriculture by offering a valuable technology that supports the goals of enhancing efficiency, mitigating methane emissions, and achieving carbon neutrality, in response to the challenges posed by climate change.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.20 no.3
• /
• pp.243-251
• /
• 2018

The gridded simulation of crop growth, which would be useful for shareholders and policy makers, often requires specialized computation tasks for preparation of weather input data and operation of a given crop model. Here we developed an automated system to allow for crop growth simulation over a region using the DSSAT (Decision Support System for Agrotechnology Transfer) model. The system consists of modules implemented using R and shell script languages. One of the modules has a functionality to create weather input files in a plain text format for each cell. Another module written in R script was developed for GIS data processing and parallel computing. The other module that launches the crop model automatically was implemented using the shell script language. As a case study, the automated system was used to determine the maximum soybean yield for a given set of management options in Illinois state in the US. The AgMERRA dataset, which is reanalysis data for agricultural models, was used to prepare weather input files during 1981 - 2005. It took 7.38 hours to create 1,859 weather input files for one year of soybean growth simulation in Illinois using a single CPU core. In contrast, the processing time decreased considerably, e.g., 35 minutes, when 16 CPU cores were used. The automated system created a map of the maturity group and the planting date that resulted in the maximum yield in a raster data format. Our results indicated that the automated system for the DSSAT model would help spatial assessments of crop yield at a regional scale.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2017.06a
• /
• pp.308-308
• /
• 2017

The low and unstable yield of soybean has been a major problem in Japan. Excess soil moisture conditions are one of the major factors to restrict soybean productivity. More than 80 % of soybean crops are cultivated in converted paddy fields which often have poor drainage. In central and eastern regions of Japan, the early vegetative growth of soybean tends to be restricted by the flooding damage because the early growth period is overlapped with the rainy season. Field observation shows that induced excess water stress in early vegetative stage reduces dry matter production by decreasing intercepted radiation by leaf and radiation use efficiency (RUE) (Bajgain et al., 2015). Therefore, it is necessary to evaluate the responses of soybean growth for excess water conditions to assess these effects on soybean productions. In this study, we aim to modify the soybean crop model (Sinclair et al., 2003) by adding the components of the restriction of leaf area development and RUE for adaptable to excess water conditions. This model was consist of five components, phenological model, leaf area development model, dry matter production model, plant nitrogen model and soil water balance model. The model structures and parameters were estimated from the data obtained from the field experiment in Tsukuba. The excess water effects on the leaf area development were modeled with consideration of decrease of blanch emergence and individual leaf expansion as a function of temperature and ground water level from pot experiments. The nitrogen fixation and nitrogen absorption from soil were assumed to be inhibited by excess water stress and the RUE was assumed to be decreasing according to the decline of leaf nitrogen concentration. The results of the modified model were better agreement with the field observations of the induced excess water stress in paddy field. By coupling the crop model and the ground water level model, it may be possible to assess the impact of excess water conditions for soybean production quantitatively.

• Korean Journal of Organic Agriculture
• /
• v.25 no.4
• /
• pp.805-819
• /
• 2017

The study was carried out to develop a mathematical model for evaluating the effect of organic fertilizers in organic rice production systems. A function to simulate the nitrogen mineralization process in the paddy soil has been developed and integrated into ORYZA2000 crop growth model. Inorganic nitrogen in the soil was estimated by single exponential models, given temperature and C:N ratio of organic amendments. Data collected from the two-year field experiment were used to evaluate the performance of the model. The revised version of ORYZA2000 provided reasonable estimates of key variables for nitrogen dynamics and crop growth in the organic rice production systems. Coefficient of determination between the measured value and simulated value were 0.6613, 0.8938, and 0.8092, respectively for soil inorganic nitrogen, total dry matter production, and rice yield. This means that the model could be used to quantify nitrogen supplying capacity of organic fertilizers relative to chemical fertilizer. Nitrogen dynamics and rice growth simulated by the model would be useful information to make decision for organic fertilization in organic rice production systems.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.6
• /
• pp.285-292
• /
• 2023

With the wide spread of smart farms and the advancements in IoT technology, it is easy to obtain additional data in addition to crop images. Consequently, deep learning-based crop disease diagnosis research utilizing multimodal data has become important. This study proposes a crop disease diagnosis method using multimodal supervised contrastive learning by expanding upon the multimodal self-supervised learning. RandAugment method was used to augment crop image and time series of environment data. These augmented data passed through encoder and projection head for each modality, yielding low-dimensional features. Subsequently, the proposed multimodal supervised contrastive loss helped features from the same class get closer while pushing apart those from different classes. Following this, the pretrained model was fine-tuned for crop disease diagnosis. The visualization of t-SNE result and comparative assessments of crop disease diagnosis performance substantiate that the proposed method has superior performance than multimodal self-supervised learning.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.4
• /
• pp.270-279
• /
• 2017

Regional assessment of crop productivity using a gridded simulation approach could aid policy making and crop management. Still, little effort has been made to develop the systems that allows gridded simulations of crop growth using ORYZA 2000 model, which has been used for predicting rice yield in Korea. The objectives of this study were to develop a series of data processing modules for creating input data files, running the crop model, and aggregating output files in a region of interest using gridded data files. These modules were implemented using C++ and R to make the best use of the features provided by these programming languages. In a case study, 13000 input files in a plain text format were prepared using daily gridded weather data that had spatial resolution of 1km and 12.5 km for the period of 2001-2010. Using the text files as inputs to ORYZA2000 model, crop yield simulations were performed for each grid cell using a scenario of crop management practices. After output files were created for grid cells that represent a paddy rice field in South Korea, each output file was aggregated into an output file in the netCDF format. It was found that the spatial pattern of crop yield was relatively similar to actual distribution of yields in Korea, although there were biases of crop yield depending on regions. It seemed that those differences resulted from uncertainties incurred in input data, e.g., transplanting date, cultivar in an area, as well as weather data. Our results indicated that a set of tools developed in this study would be useful for gridded simulation of different crop models. In the further study, it would be worthwhile to take into account compatibility to a modeling interface library for integrated simulation of an agricultural ecosystem.

• Korean Journal of Organic Agriculture
• /
• v.24 no.3
• /
• pp.299-313
• /
• 2016

This study aims to investigate the present status and factors influencing farmers' satisfaction on the crop insurance for pear. Data analyzed were collected by survey and ordered logistic model was utilized for an empirical analysis. The results demonstrate that producers who are more highly educated and have an experience to receive an educational program related to crop insurance for pear are more likely to satisfy. In addition, it is shown that sales have a negative effect on the satisfaction whereas cultivated areas have a positive relationship with it. Based on the findings, it is necessary to develop a new educational program, strengthen public relations, and support an insurance premium for improving farmers' satisfaction of the insurance for pear.

• Wind and Structures
• /
• v.5 no.2_3_4
• /
• pp.127-140
• /
• 2002

This paper presents some of the results of a project whose aim has been to produce a full simulation model which would determine the efficacy of pesticides for use by both farmers and the bio-chemical industry. The work presented here describes how crop architecture can be mathematically modelled and how the mechanics of pesticide droplet capture can be simulated so that if a wind assisted droplet-trajectory model is assumed then droplet deposition patterns on crop surfaces can be predicted. This achievement, when combined with biological response models, will then enable the efficacy of pesticide use to be predicted.

• Korean Journal of Organic Agriculture
• /
• v.25 no.2
• /
• pp.279-293
• /
• 2017

This study suggests the yield forecast models for autumn chinese cabbage and radish using crop growth and development information. For this, we construct 24 alternative yield forecast models and compare the predictive power using root mean square percentage errors. The results shows that the predictive power of model including crop growth and development informations is better than model which does not include those informations. But the forecast errors of best forecast models exceeds 5%. Thus it is important to establish reliable data and improve forecast models.

• Journal of Biosystems Engineering
• /
• v.24 no.1
• /
• pp.25-30
• /
• 1999

The effectiveness of many greenhouse environment control methodologies depends on the growth information of crops. Acquisition of the growth information of crops requires a non-invasive and continuous monitoring method. Crop growth monitoring system using digital imaging technique was developed to conduct non-destructive and intact plant growth analyses. The monitoring system automatically measures crop growth information sends an appropriate control signal to the nutrient solution supplying system. To develop the monitoring system, a linear model that explains the relationship between the fresh weight and the top projected leaf area of a lettuce plant was developed from an experiment. The monitoring system was evaluated buy successive lettuce growing experiments. Results of the experiments showed that the developed system could estimate the fresh weight of lettuce from a lettuce image by using the linear model and generate an EC control signal according to the lettuce growth stage.