Optimized Hydroponic System for Predicting Abnormal Lettuce Leaf using AI Techniques

Predicting Abnormal Leaves is very important for early crop management, to identify nutrient deficiencies or pests before they cause significant economic damage. Yet, the current technologies have constraints such as insufficient feature extraction and misclassification caused by overlapping leaf patterns. Hence, this paper introduced a novel Chimp-optimized Efficient Net prediction system (COENPS) for detecting and classifying normal and abnormal leaves. At first, the hydroponic lettuce data is collected and the noise is eliminated in the preprocessing phase. Subsequently, the informative features are selected from the preprocessed data by the chimp fitness, and further abnormal leaf is predicted and classification is accomplished as the normal and abnormal leaf. The developed COENPS models are validated in the MATLAB system and a few metrics such as accuracy, AUC, Precision, error rate, and detection time are validated. With a small error value and detection time, the newly presented COENPS model achieves high accuracy, AUC, and precision rate.