Abstract
Recently, many Deep Learning models have been employed to classify different kinds of plant diseases, but very little work has been done for disease severity detection. However, it is more important to master the severities of plant diseases accurately and timely, as it helps to make effective decisions to protect the plants from being further infected and reduce financial loss. In this paper, based on the Huanglongbing (HLB)-infected leaf images obtained from PlantVillage and crowdAI , we created a dataset with 5,406 citrus leaf images infected by HLB. Then six different kinds of popular models were trained to perform the severity detection of citrus HLB with the goal to find which types of models are more suitable to detect HLB severity with the same training circumstance. The experimental results show that the Inception_v3 model with epochs=60 can achieve higher accuracy than that of other models for severity detection with an accuracy of 74.38% due to its highly computational efficiency and small number of parameters. Additionally, aiming for evaluating whether GANs-based data augmentation can contribute to improve the model learning performance, we adopted DCGANs (Deep Convolutional Generative Adversarial Networks) to augment the original training dataset up to two times itself. Finally, a new training dataset with 14,056 leaf images composed by the original training images and the augmented ones were used to train the Inception_v3 model. As a result, we achieved an accuracy of 92.60%, about 20% higher than that of the Inception_v3 model trained by the original training dataset, which suggested that the GANs-based data augmentation is very useful to improve the model learning performance.
Highlights
In recent years, Deep Learning models are widely employed in plant disease classification and identification problems [1]–[7]
DEEP LEARNING MODELS AND TRAINING STRATEGY In [2], Brahimi et al trained six Deep Learning models, namely AlexNet, DenseNet-169, Inception v3, ResNet-34, SqueezeNet-1.1, and VGG13, for plant disease classification experiments, where three different training strategies were applied to each model
The third training strategy is called From-Scratch Strategy because it starts from random configured weights. Their experimental results show that all six models with the training strategy of Deep Strategy are better than the other training strategies, from the point of classification accuracies of plant diseases
Summary
Deep Learning models are widely employed in plant disease classification and identification problems [1]–[7]. Very little work has been done for disease severity detection. From the practical point of view, reliable, accurate and timely detection of plant disease severity is more impor-. The associate editor coordinating the review of this manuscript and approving it for publication was Quan Zou. tant for farmers comparing to disease classification, as disease severity detection is helpful to assist them to make effective decisions to protect the plants from being further infected, reduce financial loss [8], and it is beneficial to predict yield loss, monitor and forecast epidemics, assess crop germplasm for disease resistance, and better understand basic biological processes such as coevolution [9]. Inaccurate and/or imprecise disease assessments might lead to faulty decisions or probably result in severer problems.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
