Real Time Face Recognition And Tracking
Monzunur Rahman Bappy MD Masrafi
Institute Of Science Trade & Technology (ISTT) Institute of Science Trade & Technology (ISTT)
Electronics & communication Engineering (ECE) Electronics & Communication Engineering (ECE)
Mail:
[email protected] Mail:
[email protected] Abstract Result & Discussion
The face is one of the easiest ways to distinguish the individual identity We have used multiple images of same person for training the
of each other. The rapid advancements in the field of computer vision, model. Where images of same person are stored independently in
which is hacked by the portal Deep Learning algorithms made it easy to separate directories We used images of 20 persons to train the
Recognize and Identify a person based his unique facial Characteristics. model. Each directories consist of 25 images of the same person.
In a real streaming environment this concept has great importance.
The processes involved in this Face Recognition and Tracking can
This project aims at automatically calculating the time duration that a
be represented in the form of a flow chart as follows:
person spend in a room through recognition and reidemification of
person’s face in a completely automated maker by using dedicated
cameras fixed in the room entrances and exits. Tracking of visitors in
manually is time consuming and difficult task. It requires more human
intervention and time. But this project offers a complete and acute
automatic visitor tracking mechanism through deep learning. This project
can mainly consider as a way of security measure. But its scope can be
extended to several other arcos town.How to bring things in from Excel
and Word
Introduction
>Today automation is available in all fields. Here we are using a Real
Time automated system for tracking the Visitors of a room
>The manual process of recording and calculating the entry and exit
time, tracking the visitors of a room has many limitations like limited
secure, time consumption, resource requirements etc.
>The Real Time Free Recognition System is very efficient in identifying a
person based on his unique facial characteristics. This can be utilized to
calculate the time that a person spends in a room.
>This real time system will automatically monitors the presence of
individuals and automatically mark their entry and exit time, so that the
duration of time they have spend in a room can be easily and
automatically inoculated. This will increase the accuracy in computation Conclusion
and reduces the human effort in the processes.
Face recognition can be applied in Security measure at Banks,
Offices Homes etc. Face recognition has received substantial
Proposed Method attention from researches in biometrics, patted recognition field and
computer vision communities. Here Face Recognition along with real
time Tracking offers wide range of applications like Automated
Attendance System, Automated Billing based on time of
The idea is to deploy this model using a normal surveillance camera fixed
consumption, Automated Visitor Tracking, Tracking unauthorized
on the room entrances to capture the visitors faces and to calculate the
persons etc...
time that they spend on the room. The basic idea behind this project is
>Limitation:
Computer vision and the Convolutional Neural Networks (CNN). The
Image quality affects how well facial-recognition algorithms
working of the proposed system can be illustrated as follows:
work. The image quality of scanning video is quite low
compared with that of a digital camera The relative angle of the
target’s face influences the recognition score profoundly.
Extrude Enhancements
Capturing image
Capturing &
re-identifying face >Future Enhancements :
By the application of similar Real Time Face recognition and
Tracking system in future the case of anonymity can be
avoided. Nobody can remain as an anonymous person, every
body will be monitored and traceable.
Here in Phase 1 (fg1.1) the camera captures the visitor’s face and
labels it with respective timestamp. Then in Phase 2 (fig:1.2) the face
Reference
captured will be compared with the stored faces and calculate the Time 1. https
duration using the labels of matching faces. ://disruptionhub.com/5-applications-facial-recognition- technology
/
www.postersession.com
www.postersession.com
2. https://money.cnn.com/2012/01/13/technology/face_recognition/
