What question did this study set out to answer?

April 24, 2026Scientific ReportsOpen Access

A comparative analysis of single- and dual-backbone deep learning architectures with explainable AI for cherry leaf disease classification

View Full Paper

Ask AI

Bookmark

Authors

HAHüseyin Tayyip Altay ÖDÖzge Demir FEFatih Ekinci

Discussion

Member takes

Overview

This comparative analysis reveals single-backbone models outperform dual-backbone architectures in cherry leaf disease classification, suggesting clearer insights for agriculture.

Key Points

To analyze and compare the effectiveness of single- and dual-backbone deep learning architectures for classifying cherry leaf diseases.
Developed a deep learning framework for multi-class disease classification using cherry leaf images.
Evaluated multiple convolutional neural network architectures through a standardized dataset of 4,995 images across five disease categories.
Conducted statistical analysis with the Wilcoxon signed-rank test to compare accuracy and recall metrics.
ResNet50 achieved the highest classification accuracy at 98.20%, followed by EfficientNetB2 and DenseNet121.
Dual-backbone architectures reached a maximum accuracy of 97.30%, underperforming single-backbone models.
Statistical results indicated a significant overestimation of detection performance when relying solely on accuracy metrics.

Cite This Study

Altay et al. (2026) studied this question.

synapsesocial.com/papers/69eb084f553a5433e34b3535 https://doi.org/10.1038/s41598-026-50104-1

View Full Paper

Ask AI

Bookmark

April 24, 2026Scientific ReportsOpen Access

A comparative analysis of single- and dual-backbone deep learning architectures with explainable AI for cherry leaf disease classification

View Full Paper

Ask AI

Bookmark

Authors

HAHüseyin Tayyip Altay ÖDÖzge Demir FEFatih Ekinci

Discussion

Member takes

Overview

This comparative analysis reveals single-backbone models outperform dual-backbone architectures in cherry leaf disease classification, suggesting clearer insights for agriculture.

Key Points

To analyze and compare the effectiveness of single- and dual-backbone deep learning architectures for classifying cherry leaf diseases.
Developed a deep learning framework for multi-class disease classification using cherry leaf images.
Evaluated multiple convolutional neural network architectures through a standardized dataset of 4,995 images across five disease categories.
Conducted statistical analysis with the Wilcoxon signed-rank test to compare accuracy and recall metrics.
ResNet50 achieved the highest classification accuracy at 98.20%, followed by EfficientNetB2 and DenseNet121.
Dual-backbone architectures reached a maximum accuracy of 97.30%, underperforming single-backbone models.
Statistical results indicated a significant overestimation of detection performance when relying solely on accuracy metrics.

Cite This Study

Altay et al. (2026) studied this question.

synapsesocial.com/papers/69eb084f553a5433e34b3535 https://doi.org/10.1038/s41598-026-50104-1

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark