The demand for efficient water use and automatic systems has been increasing due to the frequent drought damage to crops as a result of climate change, the shortage of water resources in rural areas, and the aging of farmers. The existing automatic irrigation systems reduce the amount of labor required for irrigation and maintain soil moisture. However, the irrigation threshold criteria are user-determined as opposed to being automated according to input objectives such as improving crop productivity and saving water. In this study, an algorithm that could automatically determine suitable soil moisture according to a database and an automatic irrigation system with intermittent irrigation for efficient water use were developed. An experiment was then conducted on the productivity of crops for protected cultivation according to the application of the system. As the frequency domain reflectometry (FDR) sensor used in this system measured the volumetric water content of the soil, the soil moisture tension corresponding with the set value was converted into the volumetric water content using a regression equation. The process of intermittent irrigation was defined by using the moisture movement modeling of Hydrus 2D to reduce water loss on the soil surface and allow moisture to penetrate the soil unobstructed. An experimental field of a tomato farm was divided into empirical manual and controlled automatic irrigation plots. A total of 97.3% of the soil moisture values in the −33 kPa-controlled automatic irrigation plot and 96.6% of the soil moisture values in the −25 kPa-controlled automatic irrigation plot were within each set range during the first cropping season. During the second cropping season, a total of 94.8% of the soil moisture values in the −33 kPa-controlled automatic irrigation plot was within the set range. Compared with the empirical manual irrigation plot, the water productivity in the first cropping season was 113.9% in the −33 kPa-controlled automatic irrigation plot and 106.3% in the −25 kPa-controlled automatic irrigation plot. In the second cropping season, the water productivity was 117.3% in the −33 kPa-controlled automatic irrigation plot. Therefore, an automatic irrigation system applied with intermittent irrigation could be critical to increasing agricultural production and improving water-use efficiency.

When utilizing the Crop Water Stress Index (CWSI), the most critical factor is accurately measuring canopy temperature, which is typically done using infrared sensors and imaging cameras. In this study, however, we aimed to develop a machine learning model capable of predicting leaf temperature based on environmental data, without relying on sensors, for calculating CWSI. The data underwent preprocessing to remove outliers and missing values. The number of training data points for each factor was 307,924. After data preprocessing, a Pearson correlation analysis (bivariate correlation coefficient) was conducted to select the training data for model operation. The relationship between leaf temperature and air temperature showed a strong positive correlation of 0.928 (p < 0.01). Solar radiation and relative humidity were also found to have high correlations. However, wind speed and soil moisture tension showed very low correlations with leaf temperature and were excluded from the model operation. The Decision Tree, Random Forest, and Gradient Boosting models were selected, and each model was evaluated using RMSE (Root Mean Squared Error), MAE (Mean Absolute Error), MSE (Mean Squared Error), and R² (coefficient of determination). The evaluation results showed that the Gradient Boosting model had a high R² (0.97) and low RMSE (0.88) and MAE (0.54), making it the most suitable model for predicting leaf temperature. Through the leaf temperature prediction model developed in this study, labor and costs associated with sensors can be reduced, and by applying it to real agricultural settings, it can improve crop quality and enhance the sustainability of agriculture.