欢迎您访问《智慧农业(中英文)》官方网站! English

Smart Agriculture ›› 2020, Vol. 2 ›› Issue (3): 75-85.doi: 10.12133/j.smartag.2020.2.3.202008-SA001

• 专题--农业人工智能与大数据 • 上一篇    下一篇

基于深度学习与特征可视化方法的草地贪夜蛾及其近缘种成虫识别

魏靖1(), 王玉亭1, 袁会珠2, 张梦蕾1, 王振营2   

  1. 1.深圳市识农智能科技有限公司,广东 深圳 518063
    2.中国农业科学院 植物保护研究所,北京 100193
  • 收稿日期:2020-08-01 修回日期:2020-08-31 出版日期:2020-09-30 发布日期:2020-12-09
  • 基金资助:
    中央级公益性科研院所基本科研业务费专项(Y2019YJ06);中国农业科学院重大科研任务(CAAS-ZDRW202007)
  • 作者简介:魏 靖(1987-),男,博士,研究方向为农业害虫防治研究。E-mail:jing.wei@senseagro.com
  • 通讯作者: 王玉亭,王振营 E-mail:jing.wei@senseagro.com

Identification and Morphological Analysis of Adult Spodoptera Frugiperda and Its Close Related Species Using Deep Learning

WEI Jing1(), WANG Yuting1, YUAN Huizhu2, ZHANG Menglei1, WANG Zhenying2   

  1. 1.Shenzhen SenseAgro Technology Co. , Ltd, Shenzhen 518063, China
    2.Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, China
  • Received:2020-08-01 Revised:2020-08-31 Online:2020-09-30 Published:2020-12-09
  • corresponding author: Yuting WANG,Zhenying WANG E-mail:jing.wei@senseagro.com

摘要:

草地贪夜蛾是对粮食安全具有巨大威胁的害虫,早发现、早防治对虫情控制具有重要意义。目前,利用深度学习方法进行草地贪夜蛾及其近缘种成虫识别的相关研究存在数据量严重偏小的情况,有可能造成模型未能真正学习到草地贪夜蛾及其近缘种成虫的环形纹、肾形纹等关键视觉特征。针对上述问题,本研究在建立包含草地贪夜蛾在内的7种夜蛾科成虫,10,177幅图像组成的数据库基础上,采用迁移学习方式建立了VGG-16、ResNet-50和DenseNet-121,3种夜蛾成虫识别深度学习模型,并用相同的测试集测试了所有模型。结果表明,构建的模型识别准确率均超过了98%。此外,本研究用特征可视化技术展现了模型习得的特征,并验证了这些特征和专家进行人工识别的关键视觉特征的一致性——ResNet-50和DenseNet-121的平均特征识别率在85%左右,进一步支持了用深度学习进行草地贪夜蛾成虫实时识别的可行性。研究发现,不同模型对夜蛾科成虫视觉特征的学习能力不一样,在评价模型时不能仅看识别率,还需要加入视觉特征识别率指标对模型的学习内容进行评价。本研究通过试验证明可视化分析可以直观认识模型的特征学习情况,可为行业内或其他领域的研究人员提供参考。

关键词: 草地贪夜蛾, 夜蛾, 成虫识别, 深度学习, 视觉特征, 特征可视化, 迁移学习

Abstract:

Invasive pest fall armyworm (FAW) Spodoptera frugiperda is one of the serious threats to the food safety. Early warning and control plays a key role in FAW management. Nowadays, deep learning technology has been applied to recognize the image of FAW. However, there is a serious lack of training dataset in the current researches, which may mislead the model to learn features unrelated to the key visual characteristics (ring pattern, reniform pattern, etc.) of FAW adults and its close related species. Therefore, this research established a database of 10,177 images belonging to 7 species of noctuid adults, including FAW and 6 FAW close related species. Based on the small-scale dataset, transfer learning was used to build the recognition model of FAW adults by employing three deep learning models (VGG-16, ResNet-50 and DenseNet-121) pretrained on ImageNet. All of the models got more than 98% recognition accuracy on the same testing dataset. Moreover, by using feature visualization techniques, this research visualized the features learned by deep learning models and compared them to the related key visual characteristics recognized by human experts. The results showed that there was a high consistency between the two counterparts, i.e., the average feature recognition rate of ResNet-50 and DenseNet-121 was around 85%, which further demonstrated that it was possible to use the deep learning technology for the real-time monitoring of FAW adults. In addition, this study also found that the learning abilities of key visual characteristics among different models were different even though they have similar recognition accuracy. Herein, we suggest that when evaluating the model capacity, we should not only focus on the recognition rate, the ability of learning individual visual characteristics should be allocated importance for evaluating the model performance. For those important taxonomical traits, if the visualization results indicated that the model didn't learnt them, we should then modify our datasets or adjusting the training strategies to increase the learning ability. In conclusion, this study verified that visualizing the features learnt by the model is a good way to evaluate the learning ability of deep learning models, and to provide a possible way for other researchers in the field who want to understand the features learnt by deep learning models.

Key words: Spodoptera frugiperda, noctuid, adult moth recognition, deep learning, visual characteristics, feature visualization, transfer learning

中图分类号: