Unsupervised deep learning techniques for automatic detection of plant diseases: reducing the need of manual labelling of plant images

Journal of Mathematics in Industry

Research

Roberto Oberti¹  Alessandro Benfenati²  Paola Causin³  Giovanni Stefanello³ 

[1] Dept. of Agricultural and Environmental Sciences - Production, Landscape, Agroenergy, University of Milan, Milan, Italy;Dept. of Environmental Science and Policy, University of Milan, Milan, Italy;Dept. of Mathematics, University of Milan, Milan, Italy;

关键词: Plant disease detection;    Anomaly detection;    Unsupervised deep learning;    Powdery mildew;    Multispectral imaging;    Precision agriculture;   

DOI  :  10.1186/s13362-023-00133-6

 received in 2022-04-29, accepted in 2023-03-29,  发布年份 2023

来源: Springer

【 摘 要 】

Crop protection from diseases through applications of plant protection products is crucial to secure worldwide food production. Nevertheless, sustainable management of plant diseases is an open challenge with a major role in the economic and environmental impact of agricultural activities. A primary contribution is expected to come from precision crop protection approaches, with treatments tailored to spatial and time-specific needs of the crop, in contrast to the current practice of applying treatments uniformly to fields. In view of this, image-based automatic detection of early disease symptoms is considered a key enabling technology for high throughput scouting of the crop, in order to timely target the treatments on emerging infection spots. Thanks to the unprecedented performance in image-recognition problems, Deep Learning (DL) methods based on Convolutional Neural Networks (CNNs) have recently entered the domain of plant disease detection. This work develops two DL approaches for automatic recognition of powdery mildew disease on cucumber leaves, with a specific focus on exploring unsupervised techniques to overcome the need of large training set of manually labelled images. To this aim, autoencoder networks were implemented for unsupervised detection of disease symptoms through: i) clusterization of features in a compressed space; ii) anomaly detection. The two proposed approaches were applied to multispectral images acquired during in-vivo experiments, and the obtained results were assessed by quantitative indices. The clusterization approach showed only partially capability to provide accurate disease detection, even if it gathered some relevant information. Anomaly detection showed instead to possess a significant potential of discrimination which could be further exploited as a prior step to train more powerful supervised architectures with a very limited number of labelled samples.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

41116_2023_36_Article_IEq164.gif

41116_2023_36_Article_IEq163.gif

41116_2023_36_Article_IEq162.gif

41116_2023_36_Article_IEq161.gif

41116_2023_36_Article_IEq160.gif

41116_2023_36_Article_IEq159.gif

41116_2023_36_Article_IEq158.gif

41116_2023_36_Article_IEq157.gif

41116_2023_36_Article_IEq156.gif

41116_2023_36_Article_IEq155.gif

41116_2023_36_Article_IEq154.gif

40517_2023_256_Article_IEq75.gif

41116_2023_36_Article_IEq153.gif

41116_2023_36_Article_IEq152.gif

41116_2023_36_Article_IEq151.gif

Fig. 13

41116_2023_36_Article_IEq150.gif

40517_2023_256_Article_IEq264.gif

41116_2023_36_Article_IEq149.gif

41116_2023_36_Article_IEq148.gif

41116_2023_36_Article_IEq146.gif

Fig. 1

41116_2023_36_Article_IEq145.gif

Fig. 1

41116_2023_36_Article_IEq143.gif

41116_2023_36_Article_IEq142.gif

Fig. 8

Fig. 7

40517_2023_256_Article_IEq71.gif

40517_2023_256_Article_IEq68.gif

Fig. 1

Fig. 5

41116_2023_36_Article_IEq141.gif

41116_2023_36_Article_IEq140.gif

41116_2023_36_Article_IEq139.gif

41116_2023_36_Article_IEq134.gif

41116_2023_36_Article_IEq133.gif

41116_2023_36_Article_IEq131.gif

41116_2023_36_Article_IEq129.gif

41116_2023_36_Article_IEq128.gif

41116_2023_36_Article_IEq126.gif

41116_2023_36_Article_IEq124.gif

41116_2023_36_Article_IEq122.gif