Recognition of stripe and leaf rust disease in wheat using artificial intelligence technique
Loading...
Date
2021-10
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand)
Abstract
Stripe and leaf rust of wheat are one of the common problems not only in India but also in other wheat-growing areas, which is caused by Puccinia striiformis and Puccinia triticina respectively. Present study was conducted at Norman E. Borlaug Crop Research Center, Pantnagar, Uttarakhand in rabi season 2020-21. The investigation was undertaken to create an auto-detection model for identification of rust disease in wheat using artificial intelligence technique. Seven different deep learning approaches namely ResNet50, VGG16 without augmentation, VGG-16 with augmentation, VGG-16 with augmentation and binary classification, EfficientNetB3 with augmentation, EfficientNetB5 with augmentation and EfficientNetB7 with augmentation were used to check the best algorithm for disease identification. The classification accuracy of 56%, 68%, 70.8%, 74.1%, 69.6%, 70.8%, 71.2% and 73.4% respectively, was attained by algorithm. Automated method for an early detection of a plant disease is vital for precision crop management. The present study provides a groundwork for auto-detection of disease through smartphone. This would be beneficial to the country's farmers, who otherwise faces multiple challenges in diagnosing the disease. The study was also conducted to analyze the relation between weather parameters and leaf rust disease progression. Rate of infection was calculated which show a positively correlated with both maximum and minimum temperature and negative with relative humidity. Regression analysis was done to develop a model for predicting rate of infection, which was found to be quite accurate with R2 value of 0.637.