Classification and Identification of Weeds Using Gradient Boosting Classifiers

“Weeds present a significant challenge in agriculture, affecting crop yields and increasing reliance on herbicides. Accurate and timely weed recognition is essential for effective weed management. This study proposes a novel hybrid method that combines Convolutional Neural Networks (CNN) and Extreme Gradient Boosting Machines (XGBM) to automate weed identification. The XGBM-CNN fusion leverages the strengths of both algorithms, enhancing weed identification performance. XGBM optimizes classification by creating decision tree ensembles based on CNN-learned features. This combined approach effectively captures distinct visual characteristics of various weed species, improving differentiation from crops. Evaluation of the XGBM-CNN model demonstrates impressive performance metrics, achieving a remarkable 96.80% accuracy and an outstanding kappa coefficient of 96.70%.” These results highlight the reliability and effectiveness of this hybrid method in weed identification. The integration of XGBM and CNN facilitates robust feature extraction and precise classification, capable practical applications in precision agriculture and crop yield optimization.