Plant Disease Detection using ResNet

Deep learning and computer vision have recently emerged as useful methods for the phenotyping of sick plant tissue. The majority of prior research focused on illness categorization based on images. In conventional agricultural procedures, the diagnosis of illnesses affecting rice plants is performed by professionals in a manner that is very subjective, while laboratory testing takes a significant amount of time. As a direct result of this, there is a decrease in agricultural productivity, which results in economic loss for farmers. In order to find a solution to this problem, there is a pressing need to create methods that are quick and accurate in identifying and categorizing illnesses that affect rice plants. In the field of agriculture, the development of image-based automated systems for the categorization of rice plant diseases has become an intriguing and expanding study subject. When it comes to classifying rice plant diseases, color is one of the most crucial factors. Within the scope of this investigation, an image-based method is provided for classifying rice plant diseases based on color characteristics. Deep convolutional neural systems consume recently realized astonishing results in a number of applications, one of which is the classification of tomato plants that have been affected with many illnesses. Deep convolutional neural networks with a variety of residual networks underpin our work. In conclusion, this research study has conducted disease classification based on tomato leaves by employing a pre-trained deep CNN in conjunction with the residual network. The result that ResNet-50 produced demonstrates a remarkable result with an accuracy of 96.35%