Abstract—The Deep learning algorithms has shown promising results in solving various challenging tasks of agricultural applications. In this paper, a novel convolutional neural network based imaging solution is proposed for disease identification in Brassica Napus leaves. The state-of-the-art Convolutional Neural Network (CNN) architectures are fine-tuned for feature extraction. These features are further utilized to train SVM classifier for classification of different diseases in Brassica Napus leaves. The proposed approach is based on ensemble of two fine-tuned CNN architectures follows the hypothesis that different CNN architectures learn different feature representation from images. A dataset is created for the fine tuning of the CNN models for domain adaption. Apart from the individual CNN features, the fusion of outcomes of both the CNN models is also experimented. The experimental results show excellent performance by the individual CNN models and the fusion outperforms the individual results.
 

Index Terms—Brassica Napus, disease identification, deep learning, imaging solution

Cite: Kuldeep Singh, "A Convolutional Neural Network Based Imaging Solution for Disease Identification in Brassica Napus," Journal of Advanced Agricultural Technologies, Vol. 7, No. 2, pp. 33-37, December 2020. Doi: 10.18178/joaat.7.2.33-37

Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.