Carrot Disease Detection Using Deep Learning Approach

Abstract

Carrot is a popular nutritious vegetable and cultivated throughout the world. But Farmers still cultivating this vegetable without utilizing proper scientific technology. This may lead to financial losses as well as reduce the profit of farmers. At present, vegetable disease causes lots of financial and environmental problem. But, Early detection of vegetable disease can reduce those losses and can make farmers smile. Hence, in our research we have proposed a Deep Learning-based system for carrot disease recognition. we have experimented with healthy carrot and three common carrot disease such as: Black rot, Sclerotinia rot and Root knot. We have used Convolutional Neural Network (CNN) for feature extraction purposes and fully connected neural network (FCNN) for disease classification. Convolutional Neural Network is a great tool for image feature extraction, and it reduces the hardship of manual feature extraction. We have experimented with different convolutional model with different layer and our proposed Convolutional model gives us accuracy of almost 94%, which is certainly helpful for the farmers to identify carrot disease and maximize their benefit.