期刊文献+

基于注意力卷积神经网络的大豆害虫图像识别 被引量:25

Image recognition of soybean pests based on attention convolutional neural network
下载PDF
导出
摘要 为解决大豆害虫图像的有效识别问题,需优化传统的卷积神经网络算法。将获取到的农田大豆蚜虫图像,运用优化的卷积神经网络算法,提出一种基于注意力机制的卷积神经网络模型用于对大豆蚜虫图像的识别。通过构建基于注意力机制的卷积神经网络模型,提取出具有能够有效识别的大豆蚜虫特征,提高模型的识别准确度,达到大豆蚜虫的精确识别。识别试验结果表明:在使用相同的数据集的情况下,与普通的卷积神经网络相比,加入注意力机制的卷积神经网络对大豆蚜虫图像的识别精度更高,准确度达到96.85%,与传统的卷积神经网络相比,准确率提高6.53%。该优化算法对于大豆蚜虫的识别具有更高的识别准确率,为大豆等农作物害虫的识别提供新的思路与方法。 Convolutional neural network algorithm need to be optimizing in order to solve the problem on recognizing soybean pests images in effective way. This paper use optimized convolutional neural network algorithm based on attention mechanism model to recognizing the images of farmland soybean aphids. This paper extracted the characteristics of soybean aphids that can be effectively identified, and improved the recognition accuracy of the model and the accurate identification of soybean aphids by constructing a convolutional neural network model based on attention mechanism. The results of recognition experiment in the same database show that compared with the ordinary convolution neural network, the convolution neural network with attention mechanism has higher recognition accuracy for soybean aphid image, the accuracy is 96.85%, and the accuracy is 6.53% higher than that of the traditional convolution neural network. The optimization algorithm has higher recognition accuracy for soybean aphids, and provides a new idea and method for the identification of soybean and other crop pests.
作者 孙鹏 陈桂芬 曹丽英 Sun Peng;Chen Guifen;Cao Liying(College of Information Technology,Jilin Agricultural University,Changchun,130118,China)
出处 《中国农机化学报》 北大核心 2020年第2期171-176,共6页 Journal of Chinese Agricultural Mechanization
基金 吉林省教育厅“十三五”科研项目(JJKH20180684KJ)。
关键词 注意力机制 卷积神经网络 大豆害虫识别 attention mechanism convolutional neural network soybean aphid identification
  • 相关文献

参考文献17

二级参考文献180

共引文献460

同被引文献228

引证文献25

二级引证文献124

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部