| Abstract |
The growing need for security in different areas of human life has made biometric technologies essential for reliable identification and authentication. Among these technologies, fingerprint identification is one of the most widely used because it relies on unique patterns specific to each individual. However, traditional fingerprint identification systems face several challenges, such as handling poor-quality images, environmental variability, and vulnerability to spoofing attacks. Recently, many efficient methods have emerged, particularly those utilizing deep learning, which have made solving the problems of traditional methods easier and more effective. This progress has greatly improved fingerprint identification systems in several important ways. It has increased the accuracy of identification, reduced the time needed for processing, and enhanced the systems’ ability to prevent spoofing. These innovative approaches have enabled significant advancements in image enhancement, feature extraction, and classification accuracy, effectively addressing critical gaps in traditional systems. This survey seeks to address these gaps by providing an extensive overview of state-of-the-art methodologies used in fingerprint identification systems, with a particular focus on deep learning techniques. The current study also examines various aspects of fingerprint identification, including its applications in secure digital transactions, healthcare systems, and smart city initiatives, as well as the ethical considerations, datasets, and challenges associated with its implementation. It highlights gaps identified in previous studies and offers a thorough review of the latest methods and technologies in the field. By identifying recent trends and advancements, this study provides valuable insights that can guide future researchers in developing more effective and responsible fingerprint identification systems. © 2025 University of Bahrain. All rights reserved. |
