AUTHOR=Choudhary Manoj , Sentil Sruthi , Jones Jeffrey B. , Paret Mathews L. 

TITLE=Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2023.1292643

DOI=10.3389/fpls.2023.1292643

ISSN=1664-462X

ABSTRACT=Plant disease classification is quite complex and, in most cases, requires trained plant pathologists and sophisticated labs to accurately determine the cause. Our group for the first time used microscopic images (30X) of tomato plant diseases, for which representative plant samples were diagnostically validated to classify disease symptoms using non-coding deep learning platforms (NCDL). The mean F1 scores (s.d.) of the NCDL platforms were 98.5 (1.6) for Amazon Recognition Custom Label, 93.9 (2.5) for Clarifai, 91.6 (3.9) for Teachable Machine, 95.0 (1.9) for Google AutoML Vision, and 97.5(2.7) for Microsoft Azure Custom Vision. Accuracy of the NCDL platform for Amazon Rekognition Custom Label was 99.8% (0.2), for Clarifai was 98.7% (0.5), for Teachable Machine was 98.3% (0.4), for Google AutoML Vision was 98.9% (0.6), and for Apple CreateML was 87.3 (4.3). Upon external validation, the model's F1 score of the tested NCDL platforms dropped no more than 7%. The potential future use for these models includes development of mobile-and web-based applications for classification of plant diseases and integration with a disease management advisory system. The NCDL models also have the potential to improve early triage of symptomatic plant samples into classes that may save time in diagnostic lab sample processing.