Σύγκριση τεχνολογιών εξατομικευμένης αναγνώρισης βιογραφικών και βιομετρικών χαρακτηριστικών κατά την ταυτοποίηση
Technologies review for biographic and biometrics recognition in identification process
View/ Open
Keywords
OCRAbstract
The current thesis project takes place at a time when AI is gaining more and more ground worldwide. The production of code and applications is getting closer to what we can refer to as “machines that have the ability to learn”. Therefore, many of these applications can play an important role in fundamental functions of human society by assisting areas of social life. This thesis aims to expand a part of artificial intelligence and this part is the software technologies around optical recognition (OCR) but also its role to authentication/ authorization during login in an application. An application that uses an OCR library, named Tesseract, has been developed. Then a similar one has been developed that uses similar OCR technologies but this time it is based on the already perfected suite of the company FaceTec. Both applications seek to perform the same process, to recognize a standard public identification document and to extract some data of its owner. However, the above description is preceded by the presentation of two studies on training and the processing of OCR results. The following chapters will present the content of two research as a theory base, then the first application and the technologies of the Tesseract library will be presented, then the basic elements of technologies applied in the FaceTec suite and its locally installed application will be presented. The next step will be the comparison of the results of the two apps and the conclusions that result from this. Importance will naturally be given to the accuracy of the executions results of each application.