English

Fusarium Damaged Kernels Detection Using Transfer Learning on Deep Neural Network Architecture

Machine Learning 2018-02-02 v1 Machine Learning

Abstract

The present work shows the application of transfer learning for a pre-trained deep neural network (DNN), using a small image dataset (\approx 12,000) on a single workstation with enabled NVIDIA GPU card that takes up to 1 hour to complete the training task and archive an overall average accuracy of 94.7%94.7\%. The DNN presents a 20%20\% score of misclassification for an external test dataset. The accuracy of the proposed methodology is equivalent to ones using HSI methodology (81%91%)(81\%-91\%) used for the same task, but with the advantage of being independent on special equipment to classify wheat kernel for FHB symptoms.

Keywords

Cite

@article{arxiv.1802.00030,
  title  = {Fusarium Damaged Kernels Detection Using Transfer Learning on Deep Neural Network Architecture},
  author = {Márcio Nicolau and Márcia Barrocas Moreira Pimentel and Casiane Salete Tibola and José Mauricio Cunha Fernandes and Willingthon Pavan},
  journal= {arXiv preprint arXiv:1802.00030},
  year   = {2018}
}
R2 v1 2026-06-23T00:06:43.600Z