ABSTRACT
Plants play a major role in agriculture field, national economy, etc. Consequently, conservation of plants is essential for life survival. Similar to humans, plants are susceptible to a number of bacterial, fungal, and viral diseases due to climate. To prevent the devastation of the entire plant, early detection and treatment of these diseases are vital. In this chapter, a detection and classification model (DCM) is proposed using deep convolution neural network (DCNN) to detect and classify diseases of plants using images of the leaves. First, augmentation techniques including inverting image, flipping image, rotating image, noise injection, gamma correction, deterministic image, zooming image, and resizing image are applied to increase the dataset size, to reduce noise, and to remove unwanted background noise from images of leaves is performed. Further, augmented images are given to the convolution neural network (CNN) with multiple layers of convolution and pooling layers, which are used to identify and classify the diseased plants. Plant Village dataset is used in this experiment to train the model. After the model has been trained, it is appropriately evaluated to validate the results, and an interface is created for the early identification of plant disease and to acquaint the farmer with early prevention.
