TY - JOUR

T1 - Image-based wheat fungi diseases identification by deep learning

AU - Genaev, Mikhail A.

AU - Skolotneva, Ekaterina S.

AU - Gultyaeva, Elena I.

AU - Orlova, Elena A.

AU - Bechtold, Nina P.

AU - Afonnikov, Dmitry A.

N1 - Funding Information: The dataset preparation, annotation, and Telegram bot development were supported by RFBR grant 17-29-08028. Development of algorithms was funded by the Kurchatov Genome Center of the Institute of Cytology and Genetics of Siberian Branch of the Russian Academy of Sciences, agreement with the Ministry of Education and Science of the Russian Federation, no. 075-15-2019- 1662. The data analysis performed by the computational resources of the ?Bioinformatics? Joint Computational Center was supported by the budget project no. 0259-2021-0009.Authors are grateful to Alexey Mukhin for help in Telegram bot service develop-ment. Publisher Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland.

PY - 2021/8

Y1 - 2021/8

N2 - Diseases of cereals caused by pathogenic fungi can significantly reduce crop yields. Many cultures are exposed to them. The disease is difficult to control on a large scale; thus, one of the relevant approaches is the crop field monitoring, which helps to identify the disease at an early stage and take measures to prevent its spread. One of the effective control methods is disease identification based on the analysis of digital images, with the possibility of obtaining them in field condi-tions, using mobile devices. In this work, we propose a method for the recognition of five fungal diseases of wheat shoots (leaf rust, stem rust, yellow rust, powdery mildew, and septoria), both separately and in case of multiple diseases, with the possibility of identifying the stage of plant development. A set of 2414 images of wheat fungi diseases (WFD2020) was generated, for which expert labeling was performed by the type of disease. More than 80% of the images in the dataset correspond to single disease labels (including seedlings), more than 12% are represented by healthy plants, and 6% of the images labeled are represented by multiple diseases. In the process of creating this set, a method was applied to reduce the degeneracy of the training data based on the image hashing algorithm. The disease-recognition algorithm is based on the convolutional neural network with the EfficientNet architecture. The best accuracy (0.942) was shown by a network with a training strategy based on augmentation and transfer of image styles. The recognition method was imple-mented as a bot on the Telegram platform, which allows users to assess plants by lesions in the field conditions.

AB - Diseases of cereals caused by pathogenic fungi can significantly reduce crop yields. Many cultures are exposed to them. The disease is difficult to control on a large scale; thus, one of the relevant approaches is the crop field monitoring, which helps to identify the disease at an early stage and take measures to prevent its spread. One of the effective control methods is disease identification based on the analysis of digital images, with the possibility of obtaining them in field condi-tions, using mobile devices. In this work, we propose a method for the recognition of five fungal diseases of wheat shoots (leaf rust, stem rust, yellow rust, powdery mildew, and septoria), both separately and in case of multiple diseases, with the possibility of identifying the stage of plant development. A set of 2414 images of wheat fungi diseases (WFD2020) was generated, for which expert labeling was performed by the type of disease. More than 80% of the images in the dataset correspond to single disease labels (including seedlings), more than 12% are represented by healthy plants, and 6% of the images labeled are represented by multiple diseases. In the process of creating this set, a method was applied to reduce the degeneracy of the training data based on the image hashing algorithm. The disease-recognition algorithm is based on the convolutional neural network with the EfficientNet architecture. The best accuracy (0.942) was shown by a network with a training strategy based on augmentation and transfer of image styles. The recognition method was imple-mented as a bot on the Telegram platform, which allows users to assess plants by lesions in the field conditions.

KW - Convolutional neural network

KW - Deep learning

KW - Image recognition

KW - Leaf rust

KW - Phenotyping

KW - Powdery mildew

KW - Septoria

KW - Stem rust

KW - Wheat

KW - Yellow rust

UR - http://www.scopus.com/inward/record.url?scp=85110765982&partnerID=8YFLogxK

U2 - 10.3390/plants10081500

DO - 10.3390/plants10081500

M3 - Article

C2 - 34451545

AN - SCOPUS:85110765982

VL - 10

JO - Plants

JF - Plants

SN - 2223-7747

IS - 8

M1 - 1500

ER -