Automatic Detection and Classification of Insects Using Hybrid FF-GWO-CNN Algorithm

Pest detection in agricultural crop fields is the most challenging task, so an effective pest technique required to detect insects automatically. Image processing techniques are widely preferred science because they offer multiple advantages like maximal protection, improved management and productivity. On other hand, developing automatic monitoring system dramatically reduces workforce errors. Existing image approaches limited due disadvantages poor efficiency less accuracy. Therefore, a successful based on FF-GWO-CNN classification algorithm introduced for detection. The four-step begins with pre-processing, removing insect image’s noise sunlight illumination by utilizing adaptive median filter. insects’ size shape identified using Expectation Maximization Algorithm (EMA) clustering technique, which involves not only data but also uncovering correlations visualizing global of image. Speeded up robust feature (SURF) method employed select best possible features. Eventually, features classified introducing hybrid algorithm, combines benefits Firefly (FF), Grey Wolf Optimization (GWO) Convolutional Neural Network (CNN) enhancing entire work executed MATLAB simulation software. test result reveals that suggested has delivered optimal performance high accuracy 97.5%, precision 94%, recall 92% F-score value 92%.