Revival of Fingerprint Recognition-based Biometric Systems

Fingerprint is one of the oldest and most reliable biometric technique. This paper presents different types of fingerprint recognition techniques including contactless recognition, indoor positioning, browsing research, and liveliness detection. Contactless recognition eliminates the need of physical touch and improves user comfort and hygiene by using sophisticated approaches such as convolutional neural networks (CNNs) and minutiae-based algorithms for precise identification. Indoor positioning utilizes fingerprint to recognize an individual using machine learning methods, such k-nearest neighbors (kNN) and support vector machines (SVM). It correlates radio signals with fingerprints for accurate location monitoring. Another approach called browser fingerprint recognition identifies distinct browser settings, such that monitoring of dynamic attributes is performed. Liveness detection is used to differentiate forged fingerprint from genuine records during presentation assaults using texture analysis and deep learning algorithms. This paper presents the study of recent advancements in these domains, that emphasizes the key features and limitations of existing techniques in fingerprint-based biometric systems for future researchers.