The effect and analysis of face recognition on forensic information processes in biometric identification

In recent years, we see that information and communication technology has become widespread. This increases our probability of being a victim of cybercrime. In this age where technology and modern communication tools are spreading and multiplying, it should not be forgotten that these tools are like a sharp sword, although they are used to facilitate communication around the world. However, despite this, information and communication technology has now penetrated almost all living spaces of people on earth. In addition, the growing curiosity and accompanying dependence on information and communication technology by humans makes societies vulnerable to threats such as cybercrime committed against or through data and computer systems. Therefore, in the fight against these threats, we come across Forensic Information Science, which is included in Forensic Sciences. forensic informatics; It is not a discipline that has not yet completed its development both in Turkey and in other world countries. However, it appears as a branch of science that serves the purpose of the manifestation of justice. Responsibility for documenting and reporting incidents in a research-oriented, transparent and impartial manner in order to present the result of the process to the relevant authorities, which renews and advances itself day by day depending on technological developments, expands its service area, is obliged to carry out and finalize the forensic processes carefully and meticulously. It is a discipline. This research provides an introduction to biometric technology, especially biometric facial recognition systems, depending on the forensic discipline. Biometrics is a technology with a wide range of applications, although it has the ability to distinguish people from each other by taking advantage of their physical and behavioral characteristics. In addition, the biggest advantage of biometrics is that it has advanced security technology and high accuracy. For this reason, biometric identification systems provide important contributions to forensic informatics experts in the examination and analysis of the data collected from the crime scene, with the increasing number and type of crime involved, in clarifying the crime and identifying the criminal in a fast and accurate manner. In addition, biometrics has many functionalities and each biometric system has different variability according to its indicators. On the other hand, it should be noted that each biometric method has its own advantages and disadvantages. One of the most common and effective methods of biometrics is facial recognition. The face has unique features, indentations and protrusions. As a matter of fact, when facial features are examined, each face carries many distinctive reference points. There are about 80 reference points on the human face, expressed as nodes. These nodes are encoded with a numerical value and these codes are called face traces. When these features, called node points, are removed through face recognition systems, the identification of the criminal will be easy and possible with artificial intelligence and deep learning algorithms. In this study, two separate datasets that are open to researchers [MIT CBCL Face Recognition Database (2003-2005) and FEI Face Database Disclaimer (2005-2006)] were used. In addition, in this study, two basic face recognition methods are discussed. These methods are the photo-to-photo interrogation method for criminal investigation and the live face recognition method used in the city security management system. As a result; In order to evaluate the performance of face recognition systems in biometric identification, a number of scenarios were designed and necessary tests were carried out. The results of the analysis of the tests are presented in a comprehensive way in the findings section.