Internet of Things System for Melon/Watermelon Plant Growth with Image-Based Fruit Weight Prediction

This research develops an Internet of Things (IoT) system for watering automation and growth monitoring of melon and watermelon plants, with image processing-based fruit weight prediction. Through sensors of soil moisture, temperature, air humidity, and light intensity, the system automatically regulates watering based on environmental conditions, so that plants get water efficiently and in a timely manner. In addition, the system is equipped with a camera used to take pictures of fruits, which are then analyzed using image processing algorithms to automatically predict fruit weight. This research integrates IoT, image processing, and cloud computing technologies in one connected system, thus providing a comprehensive solution for irrigation management and crop monitoring. It is expected that this system can improve water use efficiency, reduce manual intervention, and provide more accurate crop yield predictions, thus supporting more productive and sustainable smart agriculture. This research evaluates the accuracy of soil sensors and a CNN model in agricultural monitoring, achieving high reliability with nutrient (99.9998%), pH (99.05440/0), and moisture (99.3827%) sensors, as well as fruit weight prediction (99.25% to 99.93%). These results confirm the system's suitability for precision agriculture, with ongoing developments to enhance the application interface.