GACN: Generative Adversarial Classified Network for Balancing Plant Disease Dataset and Plant Disease Recognition

Abstract

Plant diseases are a critical threat to the agricultural sector. Therefore, accurate plant disease classification is important. In recent years, some researchers have used synthetic images of GAN to enhance plant disease recognition accuracy. In this paper, we propose a generative adversarial classified network (GACN) to further improve plant disease recognition accuracy. The GACN comprises a generator, discriminator, and classifier. The proposed model can not only enhance convolutional neural network performance by generating synthetic images to balance plant disease datasets but the GACN classifier can also be directly applied to plant disease recognition tasks. Experimental results on the PlantVillage and AI Challenger 2018 datasets show that the contribution of the proposed method to improve the discriminability of the convolution neural network is greater than that of the label-conditional methods of CGAN, ACGAN, BAGAN, and MFC-GAN. The accuracy of the trained classifier for plant disease recognition is also better than that of the plant disease recognition models studied on public plant disease datasets. In addition, we conducted several experiments to observe the effects of different numbers and resolutions of synthetic images on the discriminability of convolutional neural network.

The State of the Art in Root System Architecture Image Analysis Using Artificial Intelligence: A Review

“…Another is image sample sizes or class imbalances where one or more classes have variable numbers of, or too few, sample images used for training. These types of variation can be mediated during image capture with proper planning or with the use of synthetic data as demonstrated by Wang and Cao [ 70 ]. Domain bias/shifts are challenges, and the resultant model overfitting errors occur from generalization from datasets (and image styles) not seen in training.…”

HMFN-FSL: Heterogeneous Metric Fusion Network-Based Few-Shot Learning for Crop Disease Recognition

“…Three public datasets were used in this study: Mini-ImageNet [25], Plantvillage [26], and Field-PV [27]. Plantvillage is one of the most frequently used datasets in crop leaf disease recognition studies [27][28][29][30][31], containing 50,403 images covering 38 categories for 14 crops. In addition, a field scenes disease dataset named Field-PV was selected for this study.…”