Pest detection and their classification specifically for vegetable crops is one of the most vital components of crop health and national agricultural productivity. Vegetable crops are at a high risk of pest attacks hence causing major losses and affecting food security globally. Classic approaches to pest identification that include physical inspections, or rudimentary, and mostly automatic systems, are often tedious, time-consuming, imprecise, and incapable, possibly, of identifying a pest before inflicting significant harm. To these challenges, this paper proposes a new approach of Regression Tree (RT) classification in pest detection and classification. The proposed system using RT based on more accurate and efficient by applying the modern machine learning algorithms for the identification and classification of pests. Comparison results show that the presented system has very high sensitivity, specificity, precision, and F1-score, and thus improves compared to conventional methods. The findings also show that the proposed RT-based approach improves both, the detection accuracy and the time/effort of pest management. The development of innovative approaches to pest management is a promising area that could benefit from the use of RT classification and which would help in the further development of agriculture, an increase in crop yield, and the promotion of sustainable agriculture.