Identification of rice plant diseases using lightweight attention networks

作者：

Highlights：

• Attention mechanism is embedded in mobile-nets to enhance the learning capability.

• The transfer learning is performed twice for model training.

• Loss function is optimized for multi-classification tasks and improving accuracy.

• Comparison with state-of-the-art shows the outperformance of the proposed method.

摘要

•Attention mechanism is embedded in mobile-nets to enhance the learning capability.•The transfer learning is performed twice for model training.•Loss function is optimized for multi-classification tasks and improving accuracy.•Comparison with state-of-the-art shows the outperformance of the proposed method.

论文关键词：Rice disease identification,Transfer learning,Convolutional neural networks,Image classification

论文评审过程：Received 11 August 2020, Revised 16 December 2020, Accepted 17 December 2020, Available online 5 January 2021, Version of Record 5 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114514