Pest Detection using Machine Learning

Effective disease and pest management is critical for improving crop production and quality. However, the existing disease identification techniques pose challenges to conduct precise research and analysis on time. This study proposes a novel crop pest identification approach by utilizing a multilayer network model. For comprehensive recognition and analysis, a specialized model for pest image recognition and analysis is proposed by integrating a Convolutional Neural Network (CNN) with a VGG19 transfer learning network. In the initial stage, a robust sample dataset is established for the recognition model through image data augmentation and the application of diverse operations. The proposed technique undergoes simulation analysis by using an open-source dataset. The experimental results reveal that the suggested pest recognition model attains a commendable accuracy rate of 85%. This approach holds the potential in addressing the limitations of existing methods and fostering more effective pest management strategies for achieving improved crop outcomes.