PLANT DISEASE DETECTION USING DEEP LEARNING
A CNN-based method for earlier disease detection in plants is suggested and an image of the plant's diseased portions is obtained, compared to the desired dataset, and used to predict the disease and provide subsequent treatment options.
Abstract
Food is a key aspect of our lives because it is the most fundamental necessity of all living things on our planet. Because agriculture provides the majority of our food, it is extremely significant. Agriculture works to produce food goods for the expanding population, but plant diseases also impede the growth and nutritional value of food crops. Here, we'll focus on five different plants: the tomato, the hibiscus, the spinach, the mango tree, and the bitter gourd. This paper suggests a CNN-based method for earlier disease detection in plants. The method involved the following steps: image segmentation, feature extraction, and picture pre-processing. A Convolutional Neural Network (CNN) classifier is created using the outcomes of these three phases. To conduct research and analysis, an image of the plant's diseased portions is obtained, compared to the desired dataset, and used to predict the disease and provide subsequent treatment options. Following the diagnosis of the disease, the pesticides, their quantity, and the area where they should be applied are displayed. Additionally, it will display the vicinity where pesticides can be found