Metaheuristics Feature Selection Algorithms for Identification and Classification of Mango Pests Diseases

Research in image processing and artificial intelligence (AI) is crucial for enhancing the caliber of agricultural output. A control step must be performed to increase productivity by constantly monitoring pests and associated diseases. This disease causes serious loss to leaves, flowers, and fruits. Agricultural problems are typically visible on plants, and manually detecting them requires more effort and expensive equipment. The proposed approach allows the farmer to eliminate the problems before they do any damage. The fundamental goal of this research is to apply feature selection approaches to identify and classify mango pests as effectively as possible. This research is proposed to use a particle swarm optimization (PSO) feature selection to enhance the categorization of mango pests. The results show that the proposed PSO-based feature selection approach achieves 97.60% recall, 80.50% precision, and 98.90% F-measure accuracy and exhibits improvements in the accuracy of 20.34 and 5.17% for recall, 3.22 and 2.34% for precision, and 21.90 and 7.55% over the MRMR and GA algorithms, respectively.