GreenAid: a confidence-weighted ensemble deep learning system for real-time plant disease detection and management.

Second, the ensemble’s advantage over ResNet50 alone (+0.93 pp accuracy) may appear modest, but the McNemar test (Section 4.7) confirms this difference is highly significant ( \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$p = 2 \times 10^{-5}$$\end{document} ), corresponding to approximately 120 additional correctly classified images on the test set.

also in 132,142 other papers