• Magazine of the Korean Society of Agricultural Engineers
• /
• v.40 no.1
• /
• pp.43-48
• /
• 1998

In order to provide basic information for the estimation of evapotranspiration for grass (Joycia Japonica), both field lysimeter experiment and model prediction were performed to estimate daily ET Various methods were used to predict daily reference crop ET and crop coefficients. Measured mean daily ET during the 1997 growing season was 4.5mm Model predicted mean daily ET during the 1997 growing season varied from 3.6 to 4.7mm depending on the prediction model Crop coefficients varied from 0.96 to 1.27 depending on the prediction model Comparison of the seven reference crop ET prediction methods used in this study shows that the Penman-Monteith method gave the smallest ET while the Hargreaves method gave the largest ET. The crop coefficient by the corrected Penman method was 1.03, which is closest to 1.0, suggesting that this method may he the best prediction method.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.39 no.4
• /
• pp.114-121
• /
• 1997

In order to provide basic information for the estimation of evapotranspiration in the ponded direct seeding paddy field, both field lysimeter experiment and model prediction were performed to estimate daily ET. Various methods were used to predict daily reference crop ET and crop coefficients. Measure4 mean daily ET during the 1995 growing season varied from 5.9 to 6.1 mm depending on the species, while it varied from 5.1 to 5.5 mm in 1996. Model predicted mean daily ET during the 1995 growing season varied from 3.9 to 4.9 mm depending on the prediction model, while it varied from 3.5 to 4.7 mm in 1996. The smaller ET values both measured and predicted in 1996 were caused by the low values of temperature, sunshine hours, and solar radiation. Crop coefficients varied from 1.20 to 1.50 in 1995 depending on the prediction model, while it varied from 1.10 to 1.47 in 1996. Comparison of the seven reference crop ET prediction methods used in this study shows that the Penman-Monteith method and the FAO-Radiation method gave the lowest ET while the corrected Penman method and the Hargreaves method gave the largest ET. Since crop coefficients vary to a large extent based on the prediction methods, reference crop ET prediction method should be carefully selected in irrigation planning.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 1999.08a
• /
• pp.438-446
• /
• 1999

Crop minimization of the top and bottom ends of hot rolled plate, in a plate, in a plate mill, has been investigated. The existing model to determine the edging pattern at the finishing rolling pass was not reasonable to get high width accuracy and rolling yields. New models including width prediction have been formulated by using neural network model of back propagation learning algorithm and statistical analysis based on the actual production rolling data to give the optimal pattern for minimizing trimming loss. Using these models, at a given rolling condition of broadside pass and finishing pass and the permissible condition of width variation, it was possible to minimize crip at the top and bottom ends according to optimum procedure in plate mill. An application to improve the plan view pattern reduced width variation by 23% and crop length by 30% on average with an effective fishtail crop shape.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 1986.06a
• /
• pp.30-31
• /
• 1986

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.21 no.2
• /
• pp.75-84
• /
• 2019

In this study, the random forest approach was used to predict the national mean rice yield of South Korea by using mean climatic factors at a national scale. A random forest model that used monthly climate variable and year as an important predictor in predicting crop yield. Annual yield change would be affected by technical improvement for crop management as well as climate. Year as prediction factor represent technical improvement. Thus, it is likely that the variables of importance identified for the random forest model could result in a large error in prediction of rice yield in practice. It was also found that elimination of the trend of yield data resulted in reasonable accuracy in prediction of yield using the random forest model. For example, yield prediction using the training set (data obtained from 1991 to 2005) had a relatively high degree of agreement statistics. Although the degree of agreement statistics for yield prediction for the test set (2006-2015) was not as good as those for the training set, the value of relative root mean square error (RRMSE) was less than 5%. In the variable importance plot, significant difference was noted in the importance of climate factors between the training and test sets. This difference could be attributed to the shifting of the transplanting date, which might have affected the growing season. This suggested that acceptable yield prediction could be achieved using random forest, when the data set included consistent planting or transplanting dates in the predicted area.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.235-235
• /
• 2022

Genome selection is a promising tool for plant and animal breeding, which uses genome-wide molecular marker data to capture large and small effect quantitative trait loci and predict the genetic value of selection candidates. Genomic selection has been shown previously to have higher prediction accuracies than conventional marker-assisted selection (MAS) for quantitative traits. In this study, the prediction accuracy of 10 agricultural traits in the wheat core group with 567 points was compared. We used a cross-validation approach to train and validate prediction accuracy to evaluate the effects of training population size and training model.As for the prediction accuracy according to the model, the prediction accuracy of 0.4 or more was evaluated except for the SVN model among the 6 models (GBLUP, LASSO, BayseA, RKHS, SVN, RF) used in most all traits. For traits such as days to heading and days to maturity, the prediction accuracy was very high, over 0.8. As for the prediction accuracy according to the training group, the prediction accuracy increased as the number of training groups increased in all traits. It was confirmed that the prediction accuracy was different in the training population according to the genetic composition regardless of the number. All training models were verified through 5-fold cross-validation. To verify the prediction ability of the training population of the wheat core collection, we compared the actual phenotype and genomic estimated breeding value using 35 breeding population. In fact, out of 10 individuals with the fastest days to heading, 5 individuals were selected through genomic selection, and 6 individuals were selected through genomic selection out of the 10 individuals with the slowest days to heading. Therefore, we confirmed the possibility of selecting individuals according to traits with only the genotype for a shorter period of time through genomic selection.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2017.10a
• /
• pp.8-8
• /
• 2017

• Korean Journal of Remote Sensing
• /
• v.36 no.5_1
• /
• pp.807-821
• /
• 2020

The objective of this study is to evaluate the applicability of representative spatio-temporal fusion models developed for the fusion of mid- and low-resolution satellite images in order to construct a set of time-series high-resolution images for crop monitoring. Particularly, the effects of the characteristics of input image pairs on the prediction performance are investigated by considering the principle of spatio-temporal fusion. An experiment on the fusion of multi-temporal Sentinel-2 and RapidEye images in agricultural fields was conducted to evaluate the prediction performance. Three representative fusion models, including Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM), SParse-representation-based SpatioTemporal reflectance Fusion Model (SPSTFM), and Flexible Spatiotemporal DAta Fusion (FSDAF), were applied to this comparative experiment. The three spatio-temporal fusion models exhibited different prediction performance in terms of prediction errors and spatial similarity. However, regardless of the model types, the correlation between coarse resolution images acquired on the pair dates and the prediction date was more significant than the difference between the pair dates and the prediction date to improve the prediction performance. In addition, using vegetation index as input for spatio-temporal fusion showed better prediction performance by alleviating error propagation problems, compared with using fused reflectance values in the calculation of vegetation index. These experimental results can be used as basic information for both the selection of optimal image pairs and input types, and the development of an advanced model in spatio-temporal fusion for crop monitoring.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.233-233
• /
• 2022

As the world's population grows and food needs diversify, the demand for horticultural crops for beneficial traits is increasing. In order to meet this demand, it is necessary to develop suitable cultivars and breeding methods accordingly. Breeding methods have changed over time. With the recent development of sequencing technology, the concept of genomic selection (GS) has emerged as large-scale genome information can be used. GS shows good predictive ability even for quantitative traits by using various markers, breaking away from the limitations of Marker Assisted Selection (MAS). Moreover, GS using machine learning (ML) and deep learning (DL) has been studied recently. In this study, we aim to build a system that selects phenotype-related markers using the genomic information of the pepper population and trains a genomic selection model to select individuals from the validation population. We plan to establish an optimal genome wide association analysis model by comparing and analyzing five models. Validation of molecular markers by applying linkage markers discovered through genome wide association analysis to breeding populations. Finally, we plan to establish an optimal genome selection model by comparing and analyzing 12 genome selection models. Then We will use the genome selection model of the learning group in the breeding group to verify the prediction accuracy and discover a prediction model.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.58-58
• /
• 2022

With the development of advanced technology, automation of agricultural work is spreading. In association with the 4th industrial revolution-based technology, research on field smart farm technology is being actively conducted. A state-of-the-art unmanned automated agricultural production demonstration complex was established in Naju-si, Jeollanam-do. For the operation of the demonstration area platform, it is necessary to build a sophisticated, advanced, and intelligent field smart farming model. For the operation of the unmanned automated agricultural production demonstration area platform, we are building data on the growth of soybean for smart cultivated crops and conducting research to determine the optimal time for agricultural work. In order to operate an unmanned automation platform, data is collected to discover digital factors for water management immediately after planting, water management during the growing season, and determination of harvest time. A subsurface drip irrigation system was established for smart water management. Irrigation was carried out when the soil moisture was less than 20%. For effective water management, soil moisture was measured at the surface, 15cm, and 30cm depth. Vegetation indices were collected using drones to find key factors in soybean production prediction. In addition, major growth characteristics such as stem length, number of branches, number of nodes on the main stem, leaf area index, and dry weight were investigated. By discovering digital factors for effective decision-making through data construction, it is expected to greatly enhance the efficiency of the operation of the unmanned automated agricultural production demonstration area.