Detection of Plant Diseases Using Deep Neural Network
S. Manikanda Prabu
Keywords: Deep Neural Network, Plant Disease, Segmentation, and Texture Features
Abstract
The identification and classification of plant diseases from images of the leaves is one of the most recent study issues in the field of agriculture. Utilizing image processing techniques to identify plant diseases in agricultural plants will reduce the need for farmers to take extra precautions to safeguard their crops. This paper proposes a plant detection method using deep neural network. Input images are preprocessed using image processing techniques to remove noise. Preprocessed images are subjected to segmentation process to isolate foreground from background. Texture features are computed from the segmented region. Finally, deep neural network is adopted to perform classification task. The proposed method is tested on plant village dataset. Simulation results revealed that the presented method outperformed the other models by achieving a higher classification rate.
