A Survey on Leaf Disease Detection Using Deep Learning

Abstract

Plant diseases are the leading cause of financial losses in the agriculture farming industry around the world. It is a critical factor since it reduces both the quality and the capacity of producing crops. As a result, detecting and classifying numerous plant diseases is critical, and it requires the highest attention. Fruits are an important source of nutrients in plants all over the world, but a variety of illnesses have a negative impact on fruit output and quality. As a result, the use of an effective machine vision technology not only detects but also classifies diseases in their early stages. We proposed a new grape leaf disease detection model based on generative adversarial networks. The grape sector is growing in a good way. This research offers a novel model called Leaf GAN, which is based on generative adversarial networks (GANs), to create images of four different grape leaf diseases for training identification models, focusing on the shortage of training photos of grape leaf diseases. The dense connectivity strategy and instance normalisation are fused into an efficient discriminator to identify real and fake disease images by utilising their excellent feature extraction capability on grape leaf lesions. A generator model with digressive channels is first designed to generate grape leaf disease images; then, the dense connectivity strategy and instance normalisation are fused into an efficient discriminator to identify real and fake disease images by utilising their excellent feature extraction capability on grape leaf lesions. The proposed approach is also put to the test in terms of consistency and dependability. The suggested model obtains a classification accuracy of 98.70percent after extensive simulation. The proposed work's accuracy is higher than that of typical machine learning methods.