IoT-Enabled Plant Growth Prediction and Health Monitoring System Using Sensor Fusion and Machine Learning Techniques

Abstract

The major challenges farmers face are predicting plant growth and identifying health problems before it is too late. The manual observations in old methods typically result in resource waste and erroneous predictions, damaging the ecosystem and crop production. Getting a dependable and automated system to mitigate the challenges is now more important than ever.

Given this pressing need, this paper proposes a creative solution using environmental and plant-specific sensors to collect real-time data. Then, it will be analysed using simplified machine learning algorithms, specifically Random Forest Classifiers, to precisely forecast plant growth stages and Support Vector Machine (SVM) to detect potential health problems. After testing this on various plant types, the accuracy of growth prediction was approximately 92.5% and 95.2% while detecting the plant's health issues.

This system optimises crop yields and reduces resource consumption while minimising environmental impact. Furthermore, the system is flexible and more suitable for diverse farming needs, including smart farming and managing greenhouses. This research enables the farmers to make informed decisions and cultivate a more sustainable future.