Evaluating Trust in CNN Transfer Learning with Flower Image Classification via Heatmap-Based XAI
1
Issued Date
2025-07-01
Resource Type
eISSN
22869131
Scopus ID
2-s2.0-105012378842
Journal Title
Ecti Transactions on Computer and Information Technology
Volume
19
Issue
3
Start Page
392
End Page
405
Rights Holder(s)
SCOPUS
Bibliographic Citation
Ecti Transactions on Computer and Information Technology Vol.19 No.3 (2025) , 392-405
Suggested Citation
Tanawongsuwan R., Phongsuphap S., Mongkolwat P. Evaluating Trust in CNN Transfer Learning with Flower Image Classification via Heatmap-Based XAI. Ecti Transactions on Computer and Information Technology Vol.19 No.3 (2025) , 392-405. 405. doi:10.37936/ecti-cit.2025193.260320 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/111612
Title
Evaluating Trust in CNN Transfer Learning with Flower Image Classification via Heatmap-Based XAI
Author(s)
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
Convolutional neural networks (CNNs) have demonstrated impressive performance in image classification tasks but are often criticized for their black-box nature, which complicates understanding their decision-making and reliability. Transfer learning with pre-trained CNNs is a widely used approach for tasks with limited data. This study evaluates the performance and explainability of popular CNN models on flower image classification using two custom datasets, Flower-8-One and Flower-8-Zoom. Employing Explainable AI (XAI) techniques, such as Grad-CAM, this research visualizes CNN decision-making to uncover its alignment with human perception. A human study assesses trustworthiness by analyzing participants' confidence scores based on model visualizations. Results indicate strong CNN performance but highlight disparities between model-extracted features and human expectations. Among the models evaluated, Xception and Inception-v3 consistently earn higher trust ratings. These findings emphasize the necessity of XAI-driven evaluations to enhance trust and reliability in CNN-integrated systems, particularly in applications requiring human-computer interaction.