AI-Based Plant Health Monitoring Under Stress Conditions

Living organisms adjust their physiological processes in response to change in environmental conditions using a complete signaling network. Electrical signals carry useful information and transfer it through a fast process compared to other signaling techniques. Like animals, plant health status can be measured by analyzing bio-electrical activity related signals. In this paper, we have developed a new machine learning-based algorithm to detect the physiological conditions of plants by exploring the inherent characteristics of plant electrical signals. The plant responds in a different way in response to different stress conditions. This paper studies the changes of plant response corresponding to excess water stress condition, less water stress condition, and normal water state. The experimental result of three classifiers including K-Nearest Neighbors, Naïve Bayes Classifier, and K-Means Clustering. Among these three, K-Nearest Neighbors gives the best accuracy of approximately 98% under different stress conditions. It is a useful technique to find out the status of plant health for different stress conditions before initial visual symptoms to appear.