The Codling moth (Cydia pomonella) is a major insect pest that primarily infests fruit trees, especially apple and pear trees, causing substantial global economic and agricultural issues. Researchers have explored pest classification and detection methods. With the rising popularity of the machine learning across various research areas, it’s potential for improving pest classification is increasingly recognized. This study introduces an ensemble learning-based method for identifying Codling moths through images, simplifying the categorization of orchards as either infected or uninfected. Besides reducing the workload and time of farmers, this approach can significantly mitigate Codling moth damage while preserving orchard productivity and health. In this research, the authors integrated a stacking ensemble model, amalgamating the functionalities of random forest, AdaBoost, and gradient boosting algorithms. The proposed model achieved 94% accuracy on field data. This research highlights the promise of advanced machine learning in addressing Codling moth-related challenges.