AFD-Net: Apple Foliar Disease Multi Classification using Deep Learning on Plant Pathology Dataset

Plant diseases significantly affect the crop, so their identification is very important. Correct identification of these diseases is crucial for establishing a good disease control strategy to avoid time and financial losses. In general, machines can greatly reduce the possibility of human error. In particular, computer vision techniques developed through deep learning have paved a way to detect and diagnose these plant diseases on the leaf. In this work, the model AFD-Net was developed to detect and identify various leaf diseases in apple trees. The dataset is from Kaggle 2020 and 2021 and was financially supported by the Cornell Initiative for Digital Agriculture. A AFD-Net was proposed for leaf disease classification in apple trees and the results of the efficiency of the model are compared with other state-of-the-art deep learning approaches. The results of the experiments in the validation dataset show that the proposed AFD-Net model achieves the highest values compared to other deep learning models in the original and extended datasets with 98.7% accuracy for Plant Pathology 2020 and 92.6% for Plant Pathology 2021.

Volatile Organic Compound Profiles From Wheat Diseases Are Pathogen-Specific and Can Be Exploited for Disease Classification

Plants and fungi emit volatile organic compounds (VOCs) that are either constitutively produced or are produced in response to changes in their physico-chemical status. We hypothesized that these chemical signals could be utilized as diagnostic tools for plant diseases. VOCs from several common wheat pathogens in pure culture (Fusarium graminearum, Fusarium culmorum, Fusarium avenaceum, Fusarium poae, and Parastagonospora nodorum) were collected and compared among isolates of the same fungus, between pathogens from different species, and between pathogens causing different disease groups [Fusarium head blight (FHB) and Septoria nodorum blotch (SNB)]. In addition, we inoculated two wheat varieties with either F. graminearum or P. nodorum, while one variety was also inoculated with Blumeria graminis f.sp. tritici (powdery mildew, PM). VOCs were collected 7, 14, and 21 days after inoculation. Each fungal species in pure culture emitted a different VOC blend, and each isolate could be classified into its respective disease group based on VOCs with an accuracy of 71.4 and 84.2% for FHB and SNB, respectively. When all collection times were combined, the classification of the tested diseases was correct in 84 and 86% of all cases evaluated. Germacrene D and sativene, which were associated with FHB infection, and mellein and heptadecanone, which were associated with SNB infection, were consistently emitted by both wheat varieties. Wheat plants infected with PM emitted significant amounts of 1-octen-3-ol and 3,5,5-trimethyl-2-hexene. Our study suggests that VOC blends could be used to classify wheat diseases. This is the first step toward a real-time disease detection in the field based on chemical signatures of wheat diseases.