A Hybrid CNN-Transformer Architecture with Bidirectional Cross-Attention Fusion for Efficient Flower Recognition
The Indonesian Journal of Computer Science · 29.06.2026 · 15 (3)
DOI: 10.33022/ijcs.v15i3.5181 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Indonesian Journal of Computer Science, online erschienen am 29.06.2026. Verfasst von Hersh Hama und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Automated flower recognition is crucial to the fields of agriculture and biodiversity monitoring, but deep learning models are extremely high in computing demand for resource-constrained devices. In this paper, a compact and efficient hybrid model based on EfficientNet-B4 and ViT-Small/16 is introduced. The design uses a bidirectional cross-attention fusion mechanism that uses both local edge-level features and global context to learn highly discriminative representations for fine-grained classification. The model was tested using 3,500 images from 35 species of flowers from a curated, augmented database, using 5-fold cross-validation. It outperformed the state-of-the-art architectures such as Swin Transformer and ResNet50, with 98.71% accuracy and a 98.80% F1 score with high statistical stability (95% CI:98.34%–99.09%) and with faster convergence rate. Statistically, these improvements were confirmed by pair-wise and Wilcoxon signed-rank statistical tests, which indicate the model's potential in resource-efficient real-world automated plant identification.
Zitieren
APA
Hama, H., HAMA, H., Jalal, S., Omer, S., & Ahmed, M. (2026). A Hybrid CNN-Transformer Architecture with Bidirectional Cross-Attention Fusion for Efficient Flower Recognition. The Indonesian Journal of Computer Science, 15(3). https://doi.org/10.33022/ijcs.v15i3.5181
Vancouver
Hama H, HAMA H, Jalal S, Omer S, Ahmed M. A Hybrid CNN-Transformer Architecture with Bidirectional Cross-Attention Fusion for Efficient Flower Recognition. The Indonesian Journal of Computer Science. 2026;15(3). doi:10.33022/ijcs.v15i3.5181
BibTeX
@article{Hama2026,
author = {Hama, Hersh and HAMA, Hersh and Jalal, Shayan and Omer, Saman and Ahmed, Mohammed},
title = {A Hybrid CNN-Transformer Architecture with Bidirectional Cross-Attention Fusion for Efficient Flower Recognition},
journal = {The Indonesian Journal of Computer Science},
year = {2026},
volume = {15},
number = {3},
doi = {10.33022/ijcs.v15i3.5181}
}
EUVIMED