Management of Crop Diseases through Deep Convolutional Neural Network-Based Leaf Disease Detection

Abstract

The effective management of crop leaf disease detection (CLDD) is greatly facilitated by automatic identification techniques, which have become increasingly important in agriculture. Currently, deep learning is a prominent area of research that can help manage and address the challenges faced by farmers. This paper proposes a Deep Convolutional Neural Network (DCNN) approach to manage and improve the accuracy of crop leaf disease detection. Additionally, the proposed algorithm incorporates data augmentation techniques to manage data scarcity issues arising from uneven dataset sizes. The performance of the proposed strategy is evaluated on the PlantVillage dataset for tomato plants, focusing on metrics such as accuracy, precision, recall, and F1-score, demonstrating effective management of crop disease detection. The results show significant improvements over traditional methods, highlighting the effectiveness of the proposed approach in managing crop leaf diseases.