<![CDATA[Biometric features that can be used for identification include iris, voice, DNA and fingerprints. Fingerprints are the most widely used biometric feature because of their uniqueness, universality and stability. Fingerprint recognition can be grouped into two different forms of problems, namely verification and identification. Verification is comparing one fingerprint with another fingerprint. Meanwhile, identification is matching an input fingerprint with fingerprint data in the database. Thus, identification can be interpreted as an extension of verification carried out by comparing one fingerprint to many fingerprints. Identification is inherently more complex than verification. The problem increases as the number of fingerprint datasets increases, resulting in an increase in the time required for the identification process. However, there is a way to overcome this complexity, namely classification. Apart from that, evolutionary algorithm optimization can also be carried out. The Chromosome Algorithm is an improvement on the evolutionary algorithm with a separate local search process. The memetic or chromosomal algorithm is a simple algorithm with reliable performance that can provide accurate solutions to problems in the real world. The current challenge is with the increasing growth of datasets (more than 106) which include the process of clustering text analysis, molecular DNA simulation, feature selection, and forecasting, handling large-scale optimization such as complex simulations, data mining, quantum chemistry, spectroscopic analysis, geophysical analysis, drug discovery, and fingerprint recognition studies. Chromosome algorithms have proven to be very competitive in large-scale optimization because they are based on stochastic algorithms that do not require gradient information. ]]>
Copyrights © 2024
