Dorsal Hand Vein Biometric Systems: A Survey on Methods, Applications, and Future Directions

With the increasing demand for security, biometric identification systems have gained significant importance in recent years. Traditional authentication methods (such as passwords, PINs, and access cards) suffer from security vulnerabilities, making biometric approaches more reliable alternatives. This review study focuses comprehensively on dorsal hand vein (DHV) biometrics. Due to the subcutaneous positioning of the veins, DHV patterns are highly resistant to forgery and offer high accuracy thanks to their individual uniqueness and temporal stability. The study examines key processes such as near-infrared imaging of vein patterns, preprocessing techniques, region of interest (ROI) extraction, and vein segmentation. Moreover, modern feature extraction methods, including Gabor filters, PCA, LBP-based approaches, Curvelet transform, and convolutional neural networks (CNNs), are discussed, along with matching algorithms such as Hamming distance, SVM, and KNN. Widely used datasets, methodological performance comparisons, and recent developments in feature-level fusion for multimodal biometric systems are also evaluated. The findings indicate that deep learning-based methods provide higher accuracy than traditional techniques, suggesting that DHV systems will play a prominent role in future secure authentication solutions.