Sustainable Apple Disease Management Using an Intelligent Fine-Tuned Transfer Learning-Based Model

Adel Sulaiman, Vatsala Anand, Sheifali Gupta, Hani Alshahrani (), Mana Saleh Al Reshan, Adel Rajab, Asadullah Shaikh and Ahmad Taher Azar
Additional contact information
Adel Sulaiman: Department of Computer Science, College of Computer Science and Information Systems, Najran University, Najran 66462, Saudi Arabia
Vatsala Anand: Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura 140401, Punjab, India
Sheifali Gupta: Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura 140401, Punjab, India
Hani Alshahrani: Department of Computer Science, College of Computer Science and Information Systems, Najran University, Najran 66462, Saudi Arabia
Mana Saleh Al Reshan: Department of Information Systems, College of Computer Science and Information Systems, Najran University, Najran 66462, Saudi Arabia
Adel Rajab: Department of Computer Science, College of Computer Science and Information Systems, Najran University, Najran 66462, Saudi Arabia
Asadullah Shaikh: Department of Information Systems, College of Computer Science and Information Systems, Najran University, Najran 66462, Saudi Arabia
Ahmad Taher Azar: College of Computer and Information Sciences, Prince Sultan University, Riyadh 11586, Saudi Arabia

Sustainability, 2023, vol. 15, issue 17, 1-15

Abstract: Apple foliar diseases are a group of diseases that affect the leaves of apple trees. These diseases can significantly impact apple tree health and fruit yield. Ordinary apple foliar diseases include frog_eye_leaf_spots, powdery mildew, rust, apple scabs, etc. Early detection of these diseases is important for effective apple crop management to increase the yield of apples. Therefore, this research proposes a fine-tuned EfficientNetB3 model for the quick and precise assessment of these apple foliar diseases. A dataset containing 23,187 RGB images of eleven different apple foliar diseases is used for experimentation. The proposed model is compared with four transfer learning models, i.e., InceptionResNetV2, ResNet50, AlexNet, and VGG16. All models are fine-tuned by adding different layers like the global average pooling layer, flatten layer, dropout layer, and dense layer. The performance of these five models is compared in terms of the precision, recall, accuracy, and F1-score. The EfficientNetB3 outperformed the other models in terms of all performance parameters. The best model is further optimized with the help of three optimizers, i.e., Adam, SGD, and Adagrad. The proposed model achieved the precision, recall, and F1-score values of 86%, 88%, and 86%, respectively, at 32 batch sizes and 10 epochs. This research formulated a model for an apple foliar disease diagnosis within sustainable agriculture.

Keywords: plant pathology; multi-class classification; foliar disease; optimization; EfficientNetB3 (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2023
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/15/17/13228/pdf (application/pdf)
https://www.mdpi.com/2071-1050/15/17/13228/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:15:y:2023:i:17:p:13228-:d:1232299

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().