Chanseok Ryu

Investigation into Possible Use of Methane Fermentation Digested Sludge as Liquid Fertilizer for Paddy Fields

Abstract Difference in vegetation growth, taste properties, and grain yield between liquid fertilizer (LF) and chemical fertilizer (CF) applied fields were identified and analyzed to promote the use of the methane fermentation digested sludge as LF using precision agriculture technology. Vegetation growth and these ratios of LF to CF were different at panicle initiation and heading stages but no significant difference in nitrogen content was at the heading stage. Dry mass is greater in CF fields and nitrogen concentration is higher in LF fields is confirmed. In spite of no topdressing in 2006, differences in vegetation growth ratios were decreased because of the organic nitrogen in LF. Difference in GreenNDVI was decreased at the heading stage but the pattern was not changed. Differences in taste properties were significant in 2006 but not in 2005. When grain yield would be decreased 25% by hulls, brown rice yield of LF fields in 2005 was 93% of the average amount in the region (510 kg/10a) and 84% of that (505 kg/10a) in 2006.

Estimation of Moisture Content in Cucumber and Watermelon Seedlings Using Hyperspectral Imagery

This research was conducted to estimate moisture content in cucurbitaceae seedlings, such as cucumber and watermelon, using hyperspectral imagery. Using a hyperspectral image acquisition system, the reflectance of leaf area of cucumber and watermelon seedlings was calculated after providing water stress. Then, moisture content in each seedling was measured by using a dry oven. Finally, using reflectance and moisture content, the moisture content estimation models were developed by PLSR analysis. After developing the estimation models, performance of the cucumber showed 0.73 of R, 1.45% of RMSE, and 1.58% of RE. Performance of the watermelon showed 0.66 of R, 1.06% of RMSE, and 1.14% of RE. The model performed slightly better after removing one sample from cucumber seedlings as outlier and unnecessary. Hence, the performance of new model for cucumber seedlings showed 0.79 of R, 1.10% of RMSE, and 1.20% of RE. The model performance combined with all samples showed 0.67 of R, 1.26% of RMSE, and 1.36% of RE. The model of cucumber showed better performance than the model of watermelon. This is because variables of cucumber are consisted of widely distributed variation, and it affected the performance. Further, accuracy and precision of the cucumber model were increased when an insignificant sample was eliminated from the dataset. Finally, it is considered that both models can be significantly used to estimate moisture content, as gradients of trend line are almost same and intersected. It is considered that the accuracy and precision of the estimating models possibly can be improved, if the models are constructed by using variables with widely distributed variation. The improved models will be utilized as the basis for developing low-priced sensors. Additional key words : Image processing, Non-destructive analysis, PLS-Regression model, Seedling quality

Classification of Chinese cabbage and radish based on the reflectance of hyperspectral imagery

In this research, the ground based hyperspectral reflectance of Chinese cabbage and radish depending on the vegetation growth stages was compared to each other. The classifiers namely decision tree, random forest and support vector machine were tested to check the feasibility of classification depending on the difference in hyperspectral reflectance. The ability of classifier was compared with the overall accuracy and kappa coefficient depending on the vegetation growth stages. The spectral merging was applied to find out the optimal spectral bands to make new multispectral sensor based on the commercial band pass filter with full width at half maximum (FWHM) such as 10nm, 25nm, 40nm, 50nm and 80nm. It was ascertained that the pattern of hyperspectral reflectance varied in Chinese cabbage and radish and also found a certain disparity of pattern in different vegetation growing stage. Although the classifying ability of support vector machine with linear method was higher than the other six methods, it was not suitable for new multispectral sensor. Hence, the decision tree with Rpart method is advantageous as a best classifier to make new multispectral sensor in order to separate the hyperspectral reflectance of Chinese cabbage and radish depending on the vegetation growth stages. The substantiates two alternative aggregate of bands 410nm, 430nm, 700nm and 720nm with 10nm of FWHM or 410nm, 440nm, 690nm and 720nm with 25nm of FWHM were suggested to be the best combinations to make new multispectral sensor without the overlap of FWHM.

Estimating catechin concentrations of new shoots in the green tea field using ground-based hyperspectral image

Hyperspectral camera was applied to establish the models of catechin concentration for green tea. The possibility of improvement for the models was checked by the multi-year models and the mutual prediction. ECg, EGCg and the ester catechin (ECg and EGCg) decreased with the growth but EC, EGC and the free catechin (EC and EGC) were changed by the covering. In partial least square regression (PLSR) models for each catechin, R2 (Relative Error for validation) was more than 0.785 (13.4%) for a single year data, 0.723 (13.3%) for two years data, and 0.756 (13.6%) for three years data except several catechins. It was possible to improve the precision and accuracy of models using the combination of catechin (free and ester type) or the combination of multi-year data. When each and each type of catechin model was predicted by the other year data, the accuracy of two years model improved comparing with it of a single year data. It means that the multi-year models might be more accurate than a single year models to predict the unknown data.

Estimation of Catechins Concentration of Green Tea Using Hyperspectral Remote Sensing

Abstract Models for the estimation of the concentration of catechins concentration in new green tea shoots were established using ground-based hyperspectral remote sensing. The coefficient of determination (R 2 ) was determined to be more than 0.913, the root mean squared error of prediction (RMSEP) was determined to be less than 0.617 % and the relative error of prediction (REP) was determined to be less than 6.17%, except in the EGC model (R 2 =0.512, RMSEP=0.272%, and REP=15.7%). The regression coefficients of the green, red edge and near infrared regions were all changed, indicating that those regions were important for the estimation of catechin concentration. A similar trend was noted for the regression coefficients of ECg and EGCg. Therefore, the X -loadings of the first latent variables of ECg and EGCg (ester-type catechins) and EC and EGC (free-type catechins) were compared and the similarities between each type of catechin were determind. Therefore, each type of PLS regression model was designed based on date of the ester- and free-type catechins. The accuracy of the free-type model was as follows: R 2 =0.774, RMSEP=0.273% and REP=7.85%. The accuracy of ester-type model was as follows: R 2 =0.869, RMSEP=0.991% and REP=6.99%. The regression coefficients of the free-type catechins differed from those of the ester-type catechins. Large changes to the regression coefficients of the green to red, and red edge regions were also demonstrated.

Estimation of Nitrogen Contents in Rice Plant at the Panicle Initiation Stage Using Ground-Based Hyperspectral Remote Sensing

Abstract In this study, ground-based hyperspectral remote sensing was used for estimating the nitrogen content of rice plants ( Kinu-hikari ) at the panicle initiation stage. The resulting images were separated into two parts: (1) the rice plant and (2) others (irrigation water, soil background) using the equation of “ GreenNDVI-NDVI„ . R RICE was calculated as the ratio of the reflectance for the rice plant to that for the reference board. Partial least square regression models were constructed based on the association between the reflectance of the rice plant and its nitrogen content. The precision and accuracy of the 2007 model was evaluated using the full-cross validation method, as the following results: R 2 = 0.846, RMSEP = 1.244 g/m 2 , and REP = 19.9%. The precision and accuracy of the 2008 model was evaluated as the following results: R 2 = 0.846, RMSEP = 1.049 g/m 2 , and REP = 20.2%. The 2007 model predicted RMSEP and REP values as 1.244 g/m 2 and 24.0%, respectively, for the 2008 data. Similarly, the 2008 model predicted RMSEP and REP values as 1.694 g/m 2 and 27.1%, respectively, for the 2007 data. Because of similarities in the regression coefficients of both models, in terms of precision, no considerable differences between validation and prediction results were observed. The following results were determined when precision and accuracy were estimated using full-cross validation based on two years’ data: R 2 = 0.867, RMSEP = 1.084 g/m 2 , and REP = 18.9%.

Investigation of Temporal and Spatial Variability for Green Tea Growth Using Precision Agriculture Technology

Abstract Spatial and temporal variability of new shoots (number of shoots, dry mass and nitrogen concentration) were investigated under several conditions using precision agriculture technology. The growth and spatial variability of new shoots were both determined using the normalized difference vegetation index (NDVI). At harvest, there were differences in new shoots growth depending on variety, severe shading, and nitrogen fertilizer type. There were differences in new shoot for “Ten-cya” compared to that for “Sen-cya” and temporal variability of growth had a different tendency compared to spatial variability at harvest depending on several conditions. Coefficients of determination (R 2 ) and root mean square error (RMSE) were established by the NDVI model. The accuracy was R 2 ≥0.826 with RMSE≤15.0 g/m 2 for “Sen-cya” and R 2 ≥0.877 with RMSE≤13.6 g/m 2 (vegetation coverage ratio ≤ 100%) for “Ten-cya”

Influence of Fertilizing Methane Fermentation Digested Sludge to Rice Paddy on Growth of Rice and Rice Taste

In this research, the vegetation growth and rice taste of the liquid fertilizer applied fields (LF) were compared with those of chemical fertilizer applied fields(CF) in order to confirm the possibility of methane fermentation digested sludge as liquid fertilizer using precision agriculture and remote sensing technology. In panicle initiation stage, the vegetation growth at LF was 60%~80% of it at CF and there were significant difference of nitrogen contents between CF and LF. The estimation model of nitrogen contents was established by GNDVI (R