Litcius/Paper detail

GAN based image augmentation for increased CNN performance in Paddy leaf disease classification

Shweta Lamba, Anupam Baliyan, Vinay Kukreja

20222022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE)70 citationsDOI

Abstract

A variety of fungal and bacterial leaf ailments wreak havoc on the paddy plant in the agricultural field. Early diagnosis of leaf infection can improve the yield of the crop. The modeling of an automatic disease classifier aids farmers in handling the spread of leaf disease in the agricultural field. This work cogitates three paddy leaf diseases (Bacterial blight, leaf smut, and leaf blast) for the creation of an AI-based robust detection and classification model. The dataset is collected from a variety of standard online repositories. GAN-based augmentation technique was used for increasing the size of the dataset. A novel approach to the convolutional neural network is proposed with the combination of augmentation and a CNN model tuner. The performance of CNN is evaluated in terms of accuracy achieved is 98.23\% in the classification process.

Topics & Concepts

Convolutional neural networkBlightLeaf spotComputer scienceClassifier (UML)Artificial intelligenceContextual image classificationPattern recognition (psychology)AgronomyBiologyImage (mathematics)Smart Agriculture and AIPlant Disease Management TechniquesPlant Virus Research Studies
GAN based image augmentation for increased CNN performance in Paddy leaf disease classification | Litcius