Laplacian Kernel and Deep Learning for Palmprint Classification

Sari

Palmprint classification is a robust biometric method for personal identification due to its uniqueness and stability. This study explores the use of deep learning combined with the Laplacian Kernel and Deep Morphological Processing Network (DMPN) for palmprint classification. We trained the proposed system on a dataset of palmprint images collected from 10 participants, each contributing 10 palm images. The results demonstrated that the model achieved an accuracy of 90%, with weighted precision, recall, and F1-score all at 0.9007, indicating a well-balanced classification performance. Additionally, the model achieved a weighted precision of 0.9045, emphasizing its ability to minimize false positives. The average Equal Error Rate (EER) of 0.0917 indicates an effective balance between the false acceptance rate (FAR) and false rejection rate (FRR). The system was tested under various conditions, including different orientations, lighting, and backgrounds, demonstrating its robustness in real-world scenarios. This study also compares the results with recent palmprint classification techniques, such as deep learning, GANs, and few-shot learning, and discusses potential improvements, including incorporating multi-spectral data fusion and few-shot learning to enhance performance in real-world applications.