• Sukhwinder Kaur
• Saurabh Sharma

Agricultural productivity is essential for ensuring global food security, yet it faces constant threats from plant diseases that lead to significant crop losses annually. Accurate and timely detection of plant diseases is crucial for mitigating these losses. However, traditional manual methods of disease detection are inefficient, subjective, and prone to human error. Recent advancements in deep learning and machine learning have demonstrated the potential for automating this process. This study evaluates the performance of three models for plant disease classification: a Hybrid Model built on EfficientNetB0 with added dense layers, a custom Convolutional Neural Network (CNN), and a K-Nearest Neighbors (KNN) classifier. Using the PlantVillage dataset, these models were trained and tested for classifying potato crop diseases into three categories: Potato__Early_blight, Potato__Late_blight, and Potato__healthy. The Hybrid Model achieved state-of-the-art performance with an accuracy of 99.77%, significantly surpassing the CNN and KNN models, which attained accuracies of 84.19% and 86.28%, respectively. The Hybrid Model also demonstrated near-perfect precision, recall, and F1-scores, making it a highly reliable tool for plant disease detection. The confusion matrix further confirmed the robustness of the Hybrid Model, with minimal misclassifications across all classes.

Hybrid Model, EfficientNetB0, K-NN, Precision Agriculture, Deep Learning, Transfer Learning.

"A Comparative Study of Plant Disease Classification Using a Hybrid EfficientNetB0-Based Deep Learning Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 1, page no.c435-c442, January-2025, Available :http://www.jetir.org/papers/JETIR2501252.pdf

"A Comparative Study of Plant Disease Classification Using a Hybrid EfficientNetB0-Based Deep Learning Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 1, page no. ppc435-c442, January-2025, Available at : http://www.jetir.org/papers/JETIR2501252.pdf