Yüksek LisansAçık Erişim

Data mining, artificial intelligence applications and its future in criminal i̇nvestigations real time face recognition application with python and opencv

2023
0 görüntülenme
0 i̇ndirme
Danışman: Doç. Dr. Sevil Özkınalı

Özet (EN)

Throughout history, developments in social life, economy, science and technology have affected human life in many areas, this effect has not only developed humanity in a good way, but also provided the development of humanity in the opposite direction. In addition, when the increase in the human population is added, various needs have emerged in crime investigations. The first of these is the identification of the criminals. Generally, old traditional methods that are slow in identification are used in crime investigations or biometric systems (fingerprint system, retina scanning, etc.) with data mining and artificial intelligence applications are used. In these applications, it is seen that the person needs himself, which is a different problem for us. In criminal investigations, a system that is both fast and does not require the person himself, and at the same time always has up-to-date personal data is essential. In this study, a sample application has been made with a real-time sample application with the face recognition system, in which the identity will be determined quickly without the need for the person himself. With this application, person identification is made by detecting the faces of people from a photo, video or a camera connected to the computer and recording them in the database on our computer, and then comparing this image with the face images in the database. In the application, 10 photographic data were taught to the program, 30 face data were taken for each photograph, and a data source with a total of 300 photographs was obtained. When the program was tested, (1) of (12) face data was incorrectly recognized and (1) was not recognized when it should have recognized the other, and it yielded results that could be considered successful with a correct prediction of 83.33% and an error margin of 36%. Thus, as an alternative to the traditional identification methods, a person recognition application that works online, can be controlled, can automatically identify and create identity from face images has been developed.

Yazar

Murat Yün

Bu Yayına Nasıl Atıf Yapılır

Murat Yün (Master Thesis). Data mining, artificial intelligence applications and its future in criminal i̇nvestigations real time face recognition application with python and opencv, 2023, Hitit University.

Lisans

Tüm Hakları Saklıdır

Bu eser belirtilen lisans koşulları altında paylaşılmaktadır.

Hitit University tezlerinden daha fazlası