Crop leaf disease recognition based on Self-Attention convolutional neural network

•SACNN can extract effective features of important areas for crop disease recognition.•SACNN includes Base-Net and SA network. Base-Net is used to extract global features.•SA network is used to obtain local features of the crop leaf lesion area.•Discuss the effects of SA in recognition and show the...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Computers and electronics in agriculture 2020-05, Vol.172, p.105341, Article 105341
Hauptverfasser:	Zeng, Weihui, Li, Miao
Format:	Artikel
Sprache:	eng
Schlagworte:	Accuracy Artificial neural networks Convolutional neural network Crop diseases Crop leaf disease recognition Feature extraction Image contrast Medical imaging Neural networks Object recognition Plant diseases Robustness Self-Attention (SA)
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!