| dc.description.abstract | In recent years, taking attendance by hand is a laborious and error-prone
process. It is necessary to collect attendance data in the classroom to verify the
presence of both teachers and students. This approach is inefficient, though, as it
takes a long time to call out each student's name and record their attendance.
Furthermore, by identifying their friends as present or by acting as a proxy for
someone else, students can falsify their attendance using this manual method. As a
result, an automated system that makes verifying attendance easier and gets rid of
fraud is required. The suggested approach makes use of an independent Android
application for verifying attendance in order to get around these problems. This
application utilizes various technologies to enhance the accuracy and convenience
of attendance management.
Biometric identification has gained popularity as pattern recognition problems
and are widely used in contemporary authentication systems.This is where portable
computers and mobile platforms are useful. The widespread use of smartphones that
come pre-installed with fingerprint scanners and face recognition software provide an
easily accessible platform for utilizing biometric attendance systems. Fingerprint and
facial recognition technology has the potential to develop into a trustworthy security
tool since it can recognize each individual based on their own facial structures,
minutiae features and characteristics.This was accomplished by implementing
MTCNN (Multi-task Cascaded Convolutional Networks) for face detection and a
pretrained facenet512 model for face recognition. This model analyzes the captured
images of the students and matches them with the registered faces in the
system.The application also uses Biometric API to implement fingerprint recognition
functionality on devices. Firebase Authentication, a strong authentication solution
offered by Google Firebase, was utilized for user registration and login. Attendance
information and other relevant information is stored in the Firebase Realtime
Database.
The purpose of this thesis is to implement fingerprint and facial recognition
authentication in an attendance-checking application. Combining these technologies might give a more effective and safe approach to attendance control in addition to
mitigating the drawbacks of conventional techniques. | en_US |
