Automatic Image-Based Plant Disease Severity Estimation Using Deep Learning

By Guan Wang, Yu Sun, Jianxin Wang. Computational Intelligence and Neuroscience, 2017

681 citations recorded by the source corpus.

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.

Subfields: Mechanics of Materials, Pollution

View Paper: https://www.semanticscholar.org/paper/6cc42aaf638c351a27fa052d9a5aff05de4806c6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Mechanics of Materials

Marcin Wozniak

Citations in the corpus

Identification of Maize Leaf Diseases Using Improved Deep Convolutional Neural Networks (2018, background): “The value of learning rate should be carefully evaluated [28].”
A Robust Deep-Learning-Based Detector for Real-Time Tomato Plant Diseases and Pests Recognition (2017, methodology): “Another work in the same dataset shows a test accuracy of 90.4% using a VGG-16 model trained with transfer learning [52].”
Solving Current Limitations of Deep Learning Based Approaches for Plant Disease Detection (2019, methodology): “If there is an insu ﬃ cient amount of data, transfer learning may be useful [13–15, 25 , 36– 39 ,45].”
An advanced deep learning models-based plant disease detection: A review of recent research (2023, background): “These variations can be attributed to external factors such as uneven lighting, extensive occlusion, and fuzzy details (Wang et al., 2017).”
Plant diseases and pests detection based on deep learning: a review (2021, background): “Plant diseases and pests detection in complex scenarios is a very challenging task of finegrained recognition [101].”
Automatic and Reliable Leaf Disease Detection Using Deep Learning Techniques (2021, methodology): “Guan et al. [15,16] used a pre-trained VGG16 model to predict disease severity in apple plants and achieved a 90.40% accuracy rate.”
Towards a Semantically Enriched Computational Intelligence (SECI) Framework for Smart Farming (2017, background): “For instance, in [1] a series of deep convolutional neural networks (CNN) is used to estimate disease severity from plant images.”
A review of neural networks in plant disease detection using hyperspectral data (2018, background): “Recent literature [90,156,157] is available on evaluation of deep learning models with digital photography, image analysis and hyperspectral imaging for plant disease detection.”
Plant Diseases recognition on images using Convolutional Neural Networks: A Systematic Review (2020, neutral): “The customized Data Sets are present in 66 studies (S1, S2, S3, S6, S7, S8, S9, S12, S15, S17, S18, S20, S22, S24, S25, S28, S32, S33, S34, S36 , S37, S38, S40, S43, S44, S45, S46, S47, S49, S50, S52, S58, S64, S68, S69, S70, S71, S73, S74, S75, S77, S79, S80, S82, S84 , S87, S88, S89, S90, S91, S92”
A Survey of Deep Convolutional Neural Networks Applied for Prediction of Plant Leaf Diseases (2021, methodology): “Researchers in [18,26,28,64,77,86,102] worked on detection of a single disease in plants.”
Deep Learning for Classification and Severity Estimation of Coffee Leaf Biotic Stress (2019, background): “Wang et al. (2017) proposed the use of Convolutional Networks to estimate the severity of plant diseases.”
Tomato Diseases and Pests Detection Based on Improved Yolo V3 Convolutional Neural Network (2020, background): “…to the image recognition of diseases and pests, and have made a lot of achievements in recent years (Mohanty et al., 2016; Sladojevic et al., 2016; Wang et al., 2017; Liu et al., 2018; Ferentinos, 2018; Brahimi et al., 2018; Kaur et al., 2019), especially in the disease and insect image…”
Convolutional Neural Networks for the Automatic Identification of Plant Diseases (2019, background): “In Wang et al. (2017), classes reﬂect disease severity levels.”