Automatic Image-Based Plant Disease Severity Estimation Using Deep Learning

TLDR

Automatic, accurate estimation of disease severity is essential for food security, disease management, and yield loss prediction, and deep learning offers a promising, feature‑engineering‑free approach for fine‑grained classification. The study systematically evaluates the performance of shallow networks trained from scratch versus deep models fine‑tuned by transfer learning for disease severity estimation. The authors train a series of deep convolutional neural networks on botanist‑annotated apple black rot images from the PlantVillage dataset to diagnose disease severity. The deep VGG16 model fine‑tuned by transfer learning achieved 90.4 % accuracy on a hold‑out test set, indicating strong potential for modern agricultural disease control.

Abstract

Automatic and accurate estimation of disease severity is essential for food security, disease management, and yield loss prediction. Deep learning, the latest breakthrough in computer vision, is promising for fine-grained disease severity classification, as the method avoids the labor-intensive feature engineering and threshold-based segmentation. Using the apple black rot images in the PlantVillage dataset, which are further annotated by botanists with four severity stages as ground truth, a series of deep convolutional neural networks are trained to diagnose the severity of the disease. The performances of shallow networks trained from scratch and deep models fine-tuned by transfer learning are evaluated systemically in this paper. The best model is the deep VGG16 model trained with transfer learning, which yields an overall accuracy of 90.4% on the hold-out test set. The proposed deep learning model may have great potential in disease control for modern agriculture.