Documentos de Académico
Documentos de Profesional
Documentos de Cultura
AbstractFace recognition system is widely used for human through communication port to microcontroller[3].
identification due to its capability to measure and subsequently
identifies human identification especially for security purposes. Nowadays, building security has become one of the
This paper presents the development of Graphical User Interface
important issues to deal with [2]. Many identification
(GUI) based on face recognition system and Peripheral Interface
Controller (PIC) as an input/output carrier to switch on/off processes have been introduced to improve security system for
magnetic lock for door lock security system. The development is building access control such as face recognition, iris and
implemented by interfacing GUI built in MATLABR2009a with finger identifications. Compared to others, face recognition is
microcontroller to auto-switching magnetic lock for door lock less intrusive and more secure [4]. Initially a new template
security system. USB serial communication is used to interface will be created based on users face information. The created
between GUI and PIC microcontroller that allows input data
template will be compared with the existed template in the
transmission from GUI to microcontroller. The developed system
shows that the auto-switching mode transmission being database. Access will only be granted if the matching features
implemented via PIC microcontroller and USB serial between both templates are within the set of tolerance.
communication cable. It was also found that GUI can successfully
switch on and off the magnetic lock when an authorized image Many methods can be applied in face recognition system.
from GUI database is identified. One of them is Eigenfaces. This method works when the face
images are projected onto a face space (feature space) which
Keywords Peripheral Interface Controller (PIC);
best defines as the variation of the known test images. The
microcontroller; Graphical User Interface (GUI); MATLAB;
face space is defined by the eigenfaces which are the
eigenvectors of the set of faces. Then the person identification
I. INTRODUCTION is obtained by comparing the projection of new image in
feature space with the available projection of training set
image [4]. The use of GUI can greatly simplify the process. It
The securities currently become a very important issue and
can speed up the users work, so that users no need to look up
this is why a lot of security systems have been purpose using
the command line to run the program or the system. All the
an important process such as recognition especially for
algorithms for GUI and eigenfaces recognition method are
building access controls [1]. Face recognition is one example
build using MATLABR2009a.
of process that improves the security application for building
access control. The advantage of using face recognition
Microcontrollers are used in this project because of its high
instead of other identification process such as iris or
speed performance, low-cost and programming flexibility. As
fingerprint identification is because it least intrusive and more a standard computer, the microcontroller also consists of CPU,
secure [2]. The development of GUI-based face recognition ROM, I/O ports and timer. However, they are designed to
system for door lock control using microcontroller is an execute only single specific task to control a single system.
example of security system for building access control. The Furthermore, PIC16F877A microcontroller has a built in
process start by creating a new template based on user face universal synchronous asynchronous receiver transmitter
information. This template is compared to the template exist in (USART) hardware that allows direct communication with
database, and if the template match within a set tolerance, personal computer [5]. The USART module has two modes of
access to the desire resource is granted the signal being send operations: synchronous (requires a synchronized clock
II. METHODOLOGY
V. MICROCONTROLLER PROGRAMMING
GUI window platform requires four push buttons for Figure 9. LED is turned ON as an indicator the magnetic lock turns OFF.
interfacing process. The platform allows user to select image
in training image folder as a database. Then, the user is Figure 10 shows the cases when the access is denied.
allowed to choose image in testing image folder. The system This is because the Testing image did not match with
will either identify or match both images. After the training image in the database. The magnetic door remains
identification process, the information will be sent to the close or in ON condition until the recognition system
microcontroller to switch the magnetic lock ON/OF. identify a recognize image. The LED turned OFF as an
indicator the door remains close after the image is failed to
be recognized by the system as shown in Figure 11.
Figure 11. LED is turned OFF as an indicator the magnetic lock remains to Figure 13. Overall system of GUI based face recognition system for door
turn ON. lock control using microcontroller.
VIII. CONCLUSION
Figure 12 shows the ASCII A character transmitted from
GUI to the Microcontroller. Microcontroller turned off the The GUI based face recognition system for door lock
magnetic lock when it received character A from GUI. The control using microcontroller has been developed in this study.
microcontroller compares its reference ASCII code character All related work lead to the completion of the study, including
with the data received from the GUI and switched off the the development of GUI based face recognition system and
switching transistor when the character received matched with auto-switching magnetic lock mechanism with
the character saved in microcontroller. microcontroller.
The performance and functionality of the designed system
showed that the transmission of data from GUI to
microcontroller is successfully done after the image is
recognized. The system can be improved to become fully
automatic face recognition system by adding another feature
that automatically capture image with a sensor when there is
human presence at the entrance area.
REFERENCES