Related Experiment Video
Updated: Sep 28, 2025
RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
Published on: August 8, 2017
A Hyperspectral Data 3D Convolutional Neural Network Classification Model for Diagnosis of Gray Mold Disease in
Dae-Hyun Jung1, Jeong Do Kim1, Ho-Youn Kim1
1Smart Farm Research Center, Institute of Science and Technology (KIST), Gangneung-si, South Korea.
Abstract:
Gray mold disease is one of the most frequently occurring diseases in strawberries. Given that it spreads rapidly, rapid countermeasures are necessary through the development of early diagnosis technology. In this study, hyperspectral images of strawberry leaves that were inoculated with gray mold fungus to cause disease were taken; these images were classified into healthy and infected areas as seen by the naked eye. The areas where the infection spread after time elapsed were classified as the asymptomatic class. Square regions of interest (ROIs) with a dimensionality of 16 × 16 × 150 were acquired as training data, including infected, asymptomatic, and healthy areas. Then, 2D and 3D data were used in the development of a convolutional neural network (CNN) classification model. An effective wavelength analysis was performed before the development of the CNN model. Further, the classification model that was developed with 2D training data showed a classification accuracy of 0.74, while the model that used 3D data acquired an accuracy of 0.84; this indicated that the 3D data produced slightly better performance. When performing classification between healthy and asymptomatic areas for developing early diagnosis technology, the two CNN models showed a classification accuracy of 0.73 with regards to the asymptomatic ones. To increase accuracy in classifying asymptomatic areas, a model was developed by smoothing the spectrum data and expanding the first and second derivatives; the results showed that it was possible to increase the asymptomatic classification accuracy to 0.77 and reduce the misclassification of asymptomatic areas as healthy areas. Based on these results, it is concluded that the proposed 3D CNN classification model can be used as an early diagnosis sensor of gray mold diseases since it produces immediate on-site analysis results of hyperspectral images of leaves.
Insights
Early diagnosis of strawberry gray mold is possible using a 3D convolutional neural network (CNN) model. This technology analyzes hyperspectral images for rapid, on-site detection of the disease in strawberry leaves.
Area of Science:
- Agricultural Science
- Plant Pathology
- Computer Vision
Background:
- Gray mold disease poses a significant threat to strawberry cultivation, necessitating rapid detection methods.
- Early diagnosis is crucial for timely intervention and disease management in strawberries.
Purpose of the Study:
- To develop an early diagnosis technology for strawberry gray mold disease using hyperspectral imaging and convolutional neural networks (CNNs).
- To evaluate the performance of 2D and 3D CNN models in classifying healthy, infected, and asymptomatic strawberry leaf areas.
Main Methods:
- Hyperspectral images of strawberry leaves were acquired and classified into healthy, infected, and asymptomatic categories.
- Regions of interest (ROIs) were extracted for training 2D and 3D CNN classification models.
- Effective wavelength analysis was performed, and spectral data was processed to enhance classification accuracy.
Main Results:
- The 3D CNN model achieved a higher classification accuracy (0.84) compared to the 2D CNN model (0.74).
- Classification accuracy for asymptomatic areas was improved to 0.77 after spectral data smoothing and derivative expansion.
- The developed 3D CNN model demonstrated potential for immediate on-site analysis of hyperspectral leaf images.
Conclusions:
- A 3D CNN model utilizing hyperspectral imaging is effective for the early diagnosis of strawberry gray mold.
- The proposed method offers a promising solution for rapid, on-site detection, aiding in disease management strategies.
- Enhancements to spectral data processing can further improve the accuracy of asymptomatic disease detection.
