Agricultural practitioners all hope for high yield and high quality of their crops. Measures of cultivation management are taken to improve the planting environment, fertilization, irrigation, and pest control. The agricultural practitioners get a better understanding of the cultivation environment and growth status through the analysis of collected data, which helps improve the management method. Automatic monitoring modes are set up with data analysis technologies for higher resource efficiency and benefit maximization. Based on the adaptability and output value of the crops, cultivation environment is selected either in open air or in greenhouses. High-profit crops are grown in greenhouses sheltered from external climate and kept in set conditions such as soil moisture, fertilization, temperature, luminosity, etc. with environmental control facilities (irrigation, fans, water walls, spraying, shading nets, fill lights, etc.), whose data are captured with environmental sensors. The development of Internet of Things (IoT) technology enables the wireless transmission of data to the cloud-based system. Farmers operate through mobile devices such as mobile phones and tablets, to issue commands via the network to remotely manage their farms.

Taoyuan District Agricultural Research and Extension Station, with the aim to improve the agricultural production efficiency, to reduce the crop losses caused by climate and pests and diseases, and to overcome problems of labor shortage and farming experience inheritance, has actively invested in smart agriculture projects. The Internet of Things (IoT) and AI technologies are used to integrate multiple greenhouse control facilities, set up smart agricultural development system, develop various wireless transmission environmental sensors, controller modules and facility environmental control technologies. The AI system uses deep learning to establish an image recognition model for crop growth stages and an algorithm for crop growth model using R language. It monitors the environment in real time through sensors, uses real-time image data in the field to identify and predict the growth periods, and activates management programs based on the growth periods to make adjustments on the irrigation, nutrient solutions, temperature and luminosity, to maintain precise management conditions that allow crops to grow in a stable environment with stable yields and quality. Moreover, the electronic monitoring system is developed to integrate information of geographic maps, fields, crops, digitized data, images and crop growth predictions, thereby obtaining the field management information in real time from the agricultural intelligence room.