Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206702,
        author = {Adedara Olusola Gabriel and Oke, Alice Oluwafunke and Oyinloye, Olufunke Anuoluwapo and Makinde, Oladayo and Ipeayeda, Funmilola Wumi and Falohun Adeleye Samuel},
        title = {Hyperparameter Optimization of Lightweight CNN Architectures Using Chicken Swarm Optimization for Improved Palm-Vein Biometric Performance},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {2},
        pages = {2902-2916},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206702},
        abstract = {Palm-vein biometrics recognition is suitable for secure human identification systems owing to its high discriminative capability against spoofing. This study presents an optimized palm-vein biometric recognition system using a lightweight Convolutional Neural Network enhanced with the Chicken Swarm Optimization algorithm (CSO-CNN). A total of 150 genuine and 250 impostor, palm vein samples, were employed for evaluation. The palm-vein images were subjected to standard preprocessing to enhance vein patterns prior to feature extraction. The CSO algorithm was integrated to optimize critical hyperparameters of the lightweight CNN architecture, thereby improving convergence behavior, recognition accuracy, and computational efficiency. Feature extraction was performed using the optimized CNN model, while classification was achieved using a SoftMax classifier. The system was implemented in MATLAB R2023a and optimal performance was achieved at a threshold of 0.8. Experimental results revealed that the proposed CSO-CNN significantly outperformed the conventional CNN, achieving a lower False Acceptance Rate (FAR) of 6.00%, False Rejection Rate (FRR) of 12.00%, Equal Error Rate (EER) of 10.67%, and a higher recognition accuracy of 91.75%, compared to the baseline CNN, which recorded FAR of 7.20%, FRR of 14.67%, EER of 13.33%, and accuracy of 90.00%. The optimized model reduced recognition time from 180.53 seconds to 130.52 seconds. Statistical analysis using a paired sample t-test confirmed that the observed improvement in accuracy was statistically significant (p < 0.05). These results demonstrate that CSO-based hyperparameter optimization enhances the performance of lightweight CNNs for palm-vein biometric recognition, making the proposed approach suitable for real-time and high-security authentication applications.},
        keywords = {Palm vein recognition, vascular biometrics, lightweight convolutional neural networks, hyperparameter optimization, Chicken Swarm Optimization, deep learning.},
        month = {July},
        }