Modified Metaheuristics with Transfer Learning Based Insect Pest Classification for Agricultural Crops

Abstract

Crop insect detection becomes a tedious process for agronomists because a substantial part of the crops is damaged, and due to the pest attacks, the quality is degraded. They are the major reason behind crop quality degradation and diminished crop productivity. Hence, accurate pest detection is essential to guarantee safety and crop quality. Conventional identification of insects necessitates highly trained taxonomists to detect insects precisely based on morphological features. Lately, some progress has been made in agriculture by employing machine learning (ML) to classify and detect pests. This study introduces a Modified Metaheuristics with Transfer Learning based Insect Pest Classification for Agricultural Crops (MMTL-IPCAC) technique. The presented MMTL-IPCAC technique applies contrast limited adaptive histogram equalization (CLAHE) approach for image enhancement. The neural architectural search network (NASNet) model is applied for feature extraction, and a modified grey wolf optimization (MGWO) algorithm is employed for the hyperparameter tuning process, showing the novelty of the work. At last, the extreme gradient boosting (XGBoost) model is utilized to carry out the insect classification procedure. The simulation analysis stated the enhanced performance of the MMTL-IPCAC technique in the insect classification process with maximum accuracy of 98.73%.