Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images
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 (×30) of tomato plant diseases, for which representative plant samples were diagnostica...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-01-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2023.1292643/full |
| _version_ | 1827387793464098816 |
|---|---|
| author | Manoj Choudhary Manoj Choudhary Manoj Choudhary Sruthi Sentil Jeffrey B. Jones Mathews L. Paret Mathews L. Paret |
| author_facet | Manoj Choudhary Manoj Choudhary Manoj Choudhary Sruthi Sentil Jeffrey B. Jones Mathews L. Paret Mathews L. Paret |
| author_sort | Manoj Choudhary |
| collection | DOAJ |
| description | 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 (×30) 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 (SD) of the NCDL platforms were 98.5 (1.6) for Amazon Rekognition 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. The accuracy of the NCDL platform for Amazon Rekognition Custom Label was 99.8% (0.2), for Clarifai 98.7% (0.5), for Teachable Machine 98.3% (0.4), for Google AutoML Vision 98.9% (0.6), and for Apple CreateML 87.3 (4.3). Upon external validation, the model’s accuracy of the tested NCDL platforms dropped no more than 7%. The potential future use for these models includes the development of mobile- and web-based applications for the classification of plant diseases and integration with a disease management advisory system. The NCDL models also have the potential to improve the early triage of symptomatic plant samples into classes that may save time in diagnostic lab sample processing. |
| first_indexed | 2024-03-08T16:08:26Z |
| format | Article |
| id | doaj.art-7af7f2b6bc72488ab0dcf9f128f963aa |
| institution | Directory Open Access Journal |
| issn | 1664-462X |
| language | English |
| last_indexed | 2024-03-08T16:08:26Z |
| publishDate | 2023-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj.art-7af7f2b6bc72488ab0dcf9f128f963aa2024-01-08T04:32:13ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2023-01-011410.3389/fpls.2023.12926431292643Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic imagesManoj Choudhary0Manoj Choudhary1Manoj Choudhary2Sruthi Sentil3Jeffrey B. Jones4Mathews L. Paret5Mathews L. Paret6North Florida Research and Education Center, University of Florida, Quincy, FL, United StatesPlant Pathology Department, University of Florida, Gainesville, FL, United StatesIndian Council of Agricultural Research (ICAR) - National Centre for Integrated Pest Management, New Delhi, IndiaNorth Florida Research and Education Center, University of Florida, Quincy, FL, United StatesNorth Florida Research and Education Center, University of Florida, Quincy, FL, United StatesNorth Florida Research and Education Center, University of Florida, Quincy, FL, United StatesPlant Pathology Department, University of Florida, Gainesville, FL, United StatesPlant 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 (×30) 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 (SD) of the NCDL platforms were 98.5 (1.6) for Amazon Rekognition 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. The accuracy of the NCDL platform for Amazon Rekognition Custom Label was 99.8% (0.2), for Clarifai 98.7% (0.5), for Teachable Machine 98.3% (0.4), for Google AutoML Vision 98.9% (0.6), and for Apple CreateML 87.3 (4.3). Upon external validation, the model’s accuracy of the tested NCDL platforms dropped no more than 7%. The potential future use for these models includes the development of mobile- and web-based applications for the classification of plant diseases and integration with a disease management advisory system. The NCDL models also have the potential to improve the early triage of symptomatic plant samples into classes that may save time in diagnostic lab sample processing.https://www.frontiersin.org/articles/10.3389/fpls.2023.1292643/fulldiseasescode-free modelsmachine learningtomatodeep learningbiotic stress |
| spellingShingle | Manoj Choudhary Manoj Choudhary Manoj Choudhary Sruthi Sentil Jeffrey B. Jones Mathews L. Paret Mathews L. Paret Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images Frontiers in Plant Science diseases code-free models machine learning tomato deep learning biotic stress |
| title | Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images |
| title_full | Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images |
| title_fullStr | Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images |
| title_full_unstemmed | Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images |
| title_short | Non-coding deep learning models for tomato biotic and abiotic stress classification using microscopic images |
| title_sort | non coding deep learning models for tomato biotic and abiotic stress classification using microscopic images |
| topic | diseases code-free models machine learning tomato deep learning biotic stress |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2023.1292643/full |
| work_keys_str_mv | AT manojchoudhary noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages AT manojchoudhary noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages AT manojchoudhary noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages AT sruthisentil noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages AT jeffreybjones noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages AT mathewslparet noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages AT mathewslparet noncodingdeeplearningmodelsfortomatobioticandabioticstressclassificationusingmicroscopicimages |