Detailed Plant Disease Identification Through a Neural Network and Deep Learning

Abstract

            Introduction: This research proposes a model for early and correct plant disease identification based on machine learning or, in particular, a neural network. In fact, plant diseases have posed a big threat to food security because the same leads to reduced users access to foods through decreased crop production and gain interference. Their early diagnosis becomes the key to preventing adverse health consequences for individuals. In this paper, a peculiar technique is proposed for the diagnosis of plant ailments making use of neural networks. It has two main objectives, namely to reduce or handle the problem of losses in agriculture, and social and economic impacts. Plant disease detection previously relied on a single expert view, a procedure that was time-consuming and erroneous. However, with machine learning and in particular neural networks, the revolutionizing of such diagnostic techniques was contributed. It has proposed a system that will jointly use deep learning architectures such as CNN and RNN, on the plant images together with the related disease information base replete with a large number of images for fast detection of the diseases. Thus, our study has unveiled challenges and possibilities of using neural networks in decision making regarding plant disease management. Systems like that, therefore, stand to go a long way in minimizing the social and economic impacts of crop diseases, further encouraging sustainable farming practices as well as boosting food security in light of a progressive climate.