Classification and Identification of Tomato Leaf Disease Using Deep Neural Network
Ayesha Batool, Syeda Basmah Hyder, Aymen Rahim, Namra Waheed, Muhammad Adeel Asghar, Fawad Fawad
Abstract
Agricultural productivity is something on which the economy highly depends. In addition to this, plant diseases and pests are a major problem in the agricultural sector. Their detection at the initial stage is required to get rid of all the diseases as quickly as possible and to save ourselves from the destruction of crops. Different kinds of pesticides have been used to save the plants from diseases. Even after all these safety measures, it is observed that still, the disease keeps spreading in the field. Why is it SO? The problem here arises that in many cases we are not sure of the type of disease and so a wrong pesticide might have been used instead. Hence, it all goes in vain. This means the classification of disease is as important as the detection. In this paper, an advanced classification model was proposed which detects and classifies tomato leaf disease. A training dataset consisting of 450 images is used and image features are extracted using several models and kNN is applied for the classification. Classification accuracy of 76.1% is achieved using AlexNet model and it came out to be the highest in comparison to other models.