Identification of Soybean Leaf Diseases via Deep Learning
By: Wu, Q
.
Contributor(s): Zhang, K.
Publisher: New York Springer 2019Edition: Vol.100(4), Dec.Description: 659-666p.Subject(s): Construction Engineering and Management (CEM)Online resources: Click here
In:
Journal of the institution of engineers (India): Series ASummary: We propose a novel approach for identifying soybean leaf diseases in the natural environment by convolutional neural network (CNN). AlexNet, GoogLeNet and ResNet were utilized for transfer learning. Firstly, 27 models were obtained by setting different batch sizes and the number of iterations. Then, the effects of CNN structure on identification performance were explored. The optimal model is based on ResNet and has the highest accuracy of 94.29%. In the parameter settings of the optimal network, the number of iterations and batch size are 1056 and 16, respectively, and the training depth is 140. Overall, the proposed method is effective for identifying soybean leaf diseases in the natural environment.
| Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|
Articles Abstract Database
|
School of Engineering & Technology (PG) Archieval Section | Not for loan | 2021054 |
We propose a novel approach for identifying soybean leaf diseases in the natural environment by convolutional neural network (CNN). AlexNet, GoogLeNet and ResNet were utilized for transfer learning. Firstly, 27 models were obtained by setting different batch sizes and the number of iterations. Then, the effects of CNN structure on identification performance were explored. The optimal model is based on ResNet and has the highest accuracy of 94.29%. In the parameter settings of the optimal network, the number of iterations and batch size are 1056 and 16, respectively, and the training depth is 140. Overall, the proposed method is effective for identifying soybean leaf diseases in the natural environment.
Click on an image to view it in the image viewer
There are no comments for this item.