Professional Documents
Culture Documents
Attendance Monitoring Using Facial Recognition For School Personnels
Attendance Monitoring Using Facial Recognition For School Personnels
Automated kind
Facial Recognition
1. Can the database identify similar photos entry using deep face algorithm?
Scope
Limitation
The researchers used OpenCV for Image Processing, Visual Studio 2019 and C#
programming language to make this application.
This application uses a PC or laptop to run. Users or admins can access the app
through the laptop or PC.
Development phase
The development phase shows the software and hardware that researchers used to
develop the project. Researchers used Visual studio 2019 for coding, OpenCV for
image processing.
Hardware and Software Specifications
Software
The application was developed using C# programming language and OpenCV for
image processing. The algorithm would be presented using Visual Studio 2019. The
application was developed in windows form application. So that the program would run
in windows platforms. Specifically, the researcher would demonstrate it using Windows
10 OS. The databased to be used would be stored using SQL Server 2012 and create a
folder directory for images.
Hardware
The computer hardware specifications used in the study are the following: An Acer
Aspire ES1-433 laptop with a specification of Intel(R) Core (TM) i3-6006U CPU @
2.00GHz 2.00 GHz processor, 4gigs of DDR4 for RAM, 500GB of Hard Disk Drive
and runs on 64-bit Windows 10 Education version.
Software: Simulation
Software Development Process
1. Requirement
2. Designing
3. Development
4. Testing
5. Implementation
Face Recognition Results
If the employee is in the database, his full information will appear here, including his ID, name, last
name department, and the time he entered; if he is not in the database, this information will be
unknown.
Face recognition accuracy can be measured according to the percentage of
the recognized faces per the total number of tested faces of the same person.
The researcher used equation for testing the accuracy of the application
Accuracy = (TP+TN)/(TOTAL)
Wherein:
TP = True Positive (Person within database recognized properly)
TN = True Negative (Person not in the database but correctly identified)
Dataset and Accuracy
The accuracy rate of 91% has been achieved but there are requirements or conditions to get it and it can be
even higher if the exact requirement meets
Requirements:
*Good camera quality,
*Clear or Plain background (Light Color Background)
*normal lightning or against the light
Detection Speed
We measure the speed of Detection and we have a result of 1-3 seconds detection with a good quality of angle,
movement and light. But in some cases, it takes 4-7second if the quality did not meet.
Summary
Thank you