Design of Efficient Technique for Leaf Disease Detection Using Deep Learning

  • Meeradevi, Sindhu N., Monica R. Mundada


The automatic identification and diagnosis of plant diseases are highly desired in the field of agricultural information. Deep learning is the latest topic used in pattern recognition; it can effectively solve these problems in vegetable pathology. Automatic detection using image processing techniques provide fast and accurate results. The discussed method uses LeNet model to detect diseased tomato leaves. The improved LeNet model is used for training and testing the 5 different type of tomato leaf disease and using rectified linear unit activation function and adding dropout operations and adjusting parameters the model accuracy can be improved. The experiment is run using machine learning algorithms such as J48 decision tree, decision table, Naïve Bayes classifier and comparative analysis is performed with LeNet model. It can be seen that the Convolutional Neural Networks LeNet model has the highest accuracy of 99.29% while compared to the traditional classifiers.

How to Cite
Meeradevi, Sindhu N., Monica R. Mundada. (2020). Design of Efficient Technique for Leaf Disease Detection Using Deep Learning. International Journal of Advanced Science and Technology, 29(3), 10176 - 10187. Retrieved from