Indexed In Scopus
  Scopus ID: 21100926589

CNN Based Early Detection of Leaf Disease

Dommati Vaishnavi, Thiruveedula Bharath Chandra and Elma Shrenitha

Abstract

Majority people depend on agriculture. Improper management leads to the loss of agricultural products. So, it is salient to identify the diseases that are prevalent in the leaves of plants. The term disease refers to the destruction of plants. Diseases that affect plants affect their growth. Disease detection helps reduce crop loss. This paper presents an approach for identifying the type of disease attacked by tomato plants using convolutional neural networks. Convolutional neural networks (CNN) is a subset of deep learning that is used for signal processing, picture segmentation, and other commonplace tasks like image classification, etc. Various parameters such as batch size, dropouts, and various epochs were used to assess the model’s effectiveness. The dataset contains 18345 train images and 4585 test images. CNN are used for improving model accuracy. Plants are the food source of the earth. Plant infections and diseases are therefore a major threat, the most common diagnosis consists primarily of examination of plants and presence of visual symptoms. The proposed model achieved about 95% accuracy.

Published on: December 14, 2023
doi: 10.17756/nwj.2023-s4-098
Citation: Vaishnavi D, Chandra TB, Shrenitha E. 2023. CNN Based Early Detection of Leaf Disease. NanoWorld J 9(S4): S575-S578.

Download Citation (XML)

13 Downloads
Bitnami