Automatic Detection and Classification of Healthy and Unhealthy Plant Leaves

Abstract

One of the increasing problems in the Agricultural field is the disease detection of plants. Humans get food from plants, and these are our ultimate way to live. Humans depend entirely on them, from food to eat to oxygen to breathe. They play an essential role. But nowadays, the quality of agricultural products is continuously decreasing due to various factors like plant disease such as bacteria, viruses, and fungi. If a plant is infected, it reduces quality and quantity, restricts its growth and destroys it. Because of this, there is the need for an automatic method that can help identify plant disease because manual checking of plant and identification takes more time. This approach may be extended further to various applications, as stated below. The following section presents several use cases relating to the categorization of plant leaves. Different plant leaves have other diseases. The primary concept of leaf symptoms is that bacterial, fungal, and viral illnesses predominantly impact the leaves. The breakthrough of the convergent neural networks is to automatically learn a vast number of files in parallel, directly from the training dataset, under a specific predictive modelling task like image recognition.