FACE RECOGNITION ATTENDANCE SYSTEM

CHAPTER 1
INTRODUCTION
The idea of two technologies namely Student Attendance and Feedback system has
been implemented with a machine learning approach. This system automatically detects the
student performance and maintains the student's records like attendance and their feedback on
the subjects like Science, English, etc. Automated Attendance System using Face Recognition
proposes that the system is based on face detection and recognition algorithms, which is used to
automatically detects the student face when he/she enters the class and the system is capable to
marks the attendance by recognizing him. In this proposed system the student is requested to
stand in front of the camera to detect and recognize the iris, for the system to mark attendance
for the student. Some algorithms like Gray Scale Conversion, Six Segment Rectangular Filter,
Skin Pixel Detection is being used to detect the iris.

1.1 Problem Statement

The goal of this project is to build, update and maintain an Attendance recording subsystem using
facial contours as the biometric key.

EXISTING RECOGNITION SYSTEMS:

• Fingerprint Based recognition system: In the Fingerprint based existing attendance system, a
portable fingerprint device needs to be configured with the students fingerprint earlier.
• Iris Based Recognition System: In the Iris based student attendance system, the student needs
to stand in front of a camera, so that the camera will scan the Iris of the student.
• Face Based Recognition System: The facial recognition technology can be used in recording
the attendance through a high-resolution digital camera that detects and recognizes the faces of
the students.

PROPOSED SYSTEM AND SOLUTION:

DEPT.OF CS&E, GMIT, Bharathinagara Page 1

 FACE RECOGNITION ATTENDANCE SYSTEM

The task of the proposed system is to capture the face of each student and to store it in the
database for their attendance. The face of the student needs to be captured in such a manner that all
the feature of the students' face needs are recorded and analyzed to the existing record. The main
working principle of the project is that, the video captured data is converted into image to detect and
recognize it. Further the recognized image of the student is provided with attendance, else the system
marks the database as absent.

1.2 objective:

In this project we aim to build an Attendance marking system with the help of facial
recognition owing the difficulty of the manual as well as other traditional means of attendance
systems.

1.3 Scope of the project:

Using this system, we will able to accomplish the task of marking the attendance in the classroom
automatically and output is obtained in an excel sheet as desired in real-time

• However, in order to develop a dedicated system which can be implemented in an

educational institution, a very efficient algorithm which is insensitive to the lighting
conditions of the classroom has to be developed.
• Also, a camera of the optimum resolution has to be utilized in the system.
• Another important aspect where we can work towards is creating an online database of
the attendance and automatic updating of the attendance.

CHAPTER 2

DEPT.OF CS&E, GMIT, Bharathinagara Page 2

 FACE RECOGNITION ATTENDANCE SYSTEM

LITERATURE SURVEY

2.1 Implementation of classroom attendance system based on face recognition in

class

The system consists of a camera that captures the images of the classroom and sends it to the
image enhancement module. To enhance the captured image histogram normalization, median
filtering and skin classification methods are used. Face detection is done using Viola-Jones
algorithm. Initially face detection algorithm was tested on variety of images with different face
positions and lighting conditions and then algorithm was applied to detect faces in real time video.
Algorithm is trained for the images of faces and then applied on the class room image for detection of
multiple faces in the image. The next step is face recognition, where a hybrid algorithm from PCA
and LDA is used. The detected faces are cropped from the image and compared with the face
database using an Eigen face method. The face database consists of templates of face images of
individual students that was collected and stored by an enrolment process. In this way the faces of
students are verified one by one and the attendance is marked on the server. A time table module is
attached to the system to obtain the subject, class, date and time. Teachers come in the class and just
press a button to start the attendance process.

2.2 Face Recognition-based Lecture Attendance System

The system consists of two cameras; one for determining the seating positions (fixed at the
ceiling) and the other for capturing the students face (Fixed in front of the seats). To determine the
target seat Active Student Detection (ASD) method is used to estimate the existence of a student on a
seat. One seat is targeted and camera is directed to capture the image. The face image capture is
enhanced and recognized and are recoded into the database. Every seat has a vector of values that
represent relationship between the student and seat. Attendance is estimated by interpreting the face
recognition data obtained by continuous observation. The position and attendance of the student are
recorded into the database.

2.3 Study of Implementing Automated Attendance System Using Face

Recognition Technique
The proposed system has been implemented in three basic steps. The first step is face
detection and extraction. The user stands in front of the camera and an image is captured, which is
taken as input. The frontal face is captured by using the OpenCV Har Cascade method. After the face
is detected, it is converted into a gray scale image of 50x50 pixels. The second step is to learn and
train face images. The system needs to be initialized by feeding it a set of training images of faces.
The PCA algorithm is performed on it. All the learned data is stored in an xml file. The third step is
the recognition and identification. In this step the frontal face that is to be recognized, test face, is
extracted from the image. The Eigen value for the test face is re-calculated and is matched with the

DEPT.OF CS&E, GMIT, Bharathinagara Page 3

 FACE RECOGNITION ATTENDANCE SYSTEM

stored data for the closest neighbor. Finding the closest neighbor is implemented as a function that
computes distance from the projected test face to each projected training set. The distance basis here
is "Squared Euclidean Distance." When a face is matched the corresponding information is obtained
from the database. The log table is then updated with the system time to mark the attendance of that
person.

2.4 Face Recognition Based Attendance Marking System

The system consists of a camera that must be positioned in the office room to take snap shots
of the room. These images are then sent to an enhancement module where Histogram Normalization
is used for the contrast enhancement of the image, Median Filter is used for removing noise from the
image. To avoid false detection skin classification technique is used. This process first classifies the
skin and then retains only the skin pixels and the other pixels are set to black. The enhanced image is
then sent to a face detection and recognition module. This requires MATLAB software version. Two
databases are maintained, the first one is the Face database to store the face images and extracted
features at the time of enrolment process and the second attendance database contains the information
about the employees and is also used to mark attendance.

2.5 Attendance Management System Using Face Recognition

In this system, the CCTV is fixed at the entry of the class room and is used to capture an
image of the entering student. The detected faces are stored in a database and is compared with the
existing images using Eigen faces methodology. To identify if the student image is matching, a 3D
face recognition technique is used. If a match is found, that image is processed for attendance
management. For attendance management, the attendance will be marked for the student image
matched and the information is sent to the server which controls the overall database of the student.
The software is installed in a smart phone that would help to improve the report features. When the
server receives the message of student who are absent that particular day will send an SMS to the
parent of that particular student.

CHAPTER 3

DEPT.OF CS&E, GMIT, Bharathinagara Page 4

 FACE RECOGNITION ATTENDANCE SYSTEM

METHODOLOGY
In order to obtain the attendance positions and face images in lecture, we proposed the
attendance management system based on face detection in the classroom lecture. The system
estimates the attendance and positions of each student by continuous observation and recording.
Current work is based on the method to obtain the different weights of each focused seat.

The technology aims in imparting tremendous knowledge oriented technical innovations

these days. Deep Learning is one among the interesting domain that enables the machine to train
itself by providing some datasets as input and provides an appropriate output during testing by
applying different learning algorithms. Nowadays Attendance is considered as an important factor
for both the student as well as the teacher of an educational organization. With the advancement of
the deep learning technology the machine automatically detects the attendance performance of the
students and maintains a record of those collected. In general, the attendance system of the student
can be,

1. Manual Attendance System (MAS) 2. Automated Attendance System (AAS)

1.3.1 Manual Attendance System (MAS)

Manual Student Attendance Management system is a process where a teacher concerned
with the particular attendance manually. Manual attendance may be considered as a time-
consuming process Face Recognition Based Attendance System happens for the teacher to miss
someone or students may answer multiple times on the absence of their friends. So, the problem
arises when we think about the traditional process of taking attendance in the classroom.

1.3.2 Automated Attendance System (AAS)

Automated Attendance System (AAS) is a process to student in the classroom by using face
recognition technology. It is also possible to recognize whether the student is sleeping or awake
during the lecture and it can also be implemented in the exam sessions to ensure the presence of the
student. Attendance of students will be taken by a real-time camera positioned at the door which
senses anyone entering or exiting the classroom. The camera is trained in such a way that it
differentiates shadows and photos.

The two common Human Face Recognition techniques are,

DEPT.OF CS&E, GMIT, Bharathinagara Page 5

 FACE RECOGNITION ATTENDANCE SYSTEM

1. Feature-Based approach 2. Brightness Based approach

1.3.2.1 Feature-Based approach

The Feature-based approach also known as local face recognition system, used in pointing
the key features of the face.

1.3.2.2 Brightness Based approach

Brightness-based approach also termed as the global face recognition system, used in
recognizing all the parts of the image.

CHAPTER 4

DEPT.OF CS&E, GMIT, Bharathinagara Page 6

 FACE RECOGNITION ATTENDANCE SYSTEM

SYSTEM REQUIREMENTS SPECIFICATION

4.1 Hardware Requirements

 Raspberry Pi either model A or B

(Running Raspbian OS)
 Raspberry Pi camera
 Push button
 Servo motor
 Power source

4.2 Software Requirements

 Python: generally, the language of choice for machine learning today.

 TensorFlow: an open-source machine learning and neural network toolkit. TensorFlow is

the go-to library for numerical computation and large-scale machine learning.
 scikit-learn: Simple and efficient tools for data mining and data analysis.

 scipy: a free and open-source library for scientific and technical computing.
 numpy: a Python library supporting large, multi-dimensional arrays with a large library of
functions for operating on these arrays.
 OpenCV: an open-source library of functions aimed at real-time computer vision.
 To ensure fast read and write speeds a non-traditional end program embedded RDMS,
sqlite3 was used as opposed to a traditional client-server-based database.

CHAPTER 5

DEPT.OF CS&E, GMIT, Bharathinagara Page 7

 FACE RECOGNITION ATTENDANCE SYSTEM

SYSTEM ANALYSIS AND DESIGN

5.1 Data Flow Diagram

Fig 5.1: Data Flow Diagram

5.2 Proposed System

DEPT.OF CS&E, GMIT, Bharathinagara Page 8

 FACE RECOGNITION ATTENDANCE SYSTEM

In order to obtain the attendance, positions and face images in lecture, we proposed
the attendance management system based on face detection in the classroom lecture. The system
estimates the attendance and positions of each student by continuous observation and recording.
Current work is based on the method to obtain a Realtime face recognition by placing a camera at
the entrance of the classroom. In this project we aim to build an Attendance marking system with the
help of facial recognition owing the difficulty of the manual as well as other traditional means of
attendance systems Building a solution that can be implemented in existing IoT set ups. Making the
solution highly robust and scalable. A physically connected camera system will be used to acquire
video footage from which frames containing facial information will be extracted. The acquired
images will then be pre-processed first, then fed to a facial detection algorithm for determination of
facial features. The identification algorithm will then be used to determine the identity of the
captured face with a pre-existing database.
 This project uses the Local Binary Patterns Histograms (LBPH) in OpenCV to perform
face recognition.
 Local Binary Pattern (LBP) is a simple yet very efficient texture operator which
labels the pixels of an image by thresholding the neighborhood of each pixel and
considers the result as a binary number.

 LBPH algorithm is based on 4 parameters,

 Radius: the radius is used to build the circular local binary pattern and
represents the radius around the central pixel. It is usually set to 1.
 Neighbors: the number of sample points to build the circular local binary
pattern. Keep in mind: the more sample points you include, the higher the
computational cost. It is usually set to 8.
 Grid X: the number of cells in the horizontal direction. The more cells, the finer
the grid, the higher the dimensionality of the resulting feature vector. It is
usually set to 8.
 Grid Y: the number of cells in the vertical direction. The more cells, the finer
the grid, the higher the dimensionality of the resulting feature vector. It is
usually set to 8.
 The first computational step of the LBPH is to create an intermediate image that
describes the original image in a better way, by highlighting the facial characteristics.
To do so, the algorithm uses a concept of a sliding window, based on the parameter’s
radius and neighbors.

Suppose we have a facial image in grayscale,

DEPT.OF CS&E, GMIT, Bharathinagara Page 9

 FACE RECOGNITION ATTENDANCE SYSTEM

 We can get part of this image as a window of 3x3 pixels.

 It can also be represented as a 3x3 matrix containing the intensity

of each pixel (0~255).

 Then, we need to take the central value of the matrix to be used as

the threshold.

 This value will be used to define the new values from the 8
neighbours.

 For each neighbour of the central value (threshold), we set a

new binary value. We set 1 for values equal or higher than the
threshold and 0 for values lower than the threshold.

 Now, the matrix will contain only binary values (ignoring the
central value). We need to concatenate each binary value from
each position from the matrix line by line into a new binary value
(e.g., 10001101). Note: some authors use other approaches to
concatenate the binary values (e.g., clockwise direction), but the
final result will be the same.

 Then, we convert this binary value to a decimal value and set it to

the central value of the matrix, which is actually a pixel from the
original image.

 At the end of this procedure (LBP procedure), we have a new

image which represents better the characteristics of the original
image.

DEPT.OF CS&E, GMIT, Bharathinagara Page 10

 FACE RECOGNITION ATTENDANCE SYSTEM

Figure 4.5 - Generating Binary Patterns

 As we have an image in grayscale, each histogram (from each grid) will contain only 256
positions (0~255) representing the occurrences of each pixel intensity.
 Then, we need to concatenate each histogram to create a new and bigger histogram.
Supposing we have 8x8 grids, we will have 8x8x256=16.384 positions in the final
histogram. The final histogram represents the characteristics of the image original image.

Figure 4.6 – Histograms

DEPT.OF CS&E, GMIT, Bharathinagara Page 11

 FACE RECOGNITION ATTENDANCE SYSTEM

CHAPTER 6

RESULTS
6.1 Database Before Attendance

Figure 6.1 Database Before Attendance

DEPT.OF CS&E, GMIT, Bharathinagara Page 12

 FACE RECOGNITION ATTENDANCE SYSTEM

6.2 Dataset

Figure 6.2 - Dataset

DEPT.OF CS&E, GMIT, Bharathinagara Page 13

 FACE RECOGNITION ATTENDANCE SYSTEM

6.3 Database After Attendance

Fig 6.3 Database After Attendance

DEPT.OF CS&E, GMIT, Bharathinagara Page 14

 FACE RECOGNITION ATTENDANCE SYSTEM

CHAPTER 6

TESTING

6.1Testing:
Sl.No Action Inputs Expected Actual Output Test Result
Output

1 Capture Images A Person’s Images are Images are Pass

Face Captured and captured and
Stored stored
2 Train the image Stored Create Histograms are Pass
Dataset images of a Histograms created and
face and store values are
values stored
3 Face A live stream Name of Name of Pass
Recognition of a person’s detected person detected person
face is displayed on is displayed on
the screen the screen

4 Update Multiple faces Update Attendance is Fail

attendance for from a live Attendance Updates only
multiple people video stream for all faces for a single face
at once detected

5 Detect more than 7 people Detect all 7 Only 5 faces Fail

7 faces facing the faces facing are detected at
camera the camera a time

Table 6.1: Testing

DEPT.OF CS&E, GMIT, Bharathinagara Page 15

 FACE RECOGNITION ATTENDANCE SYSTEM

6.2 Testing Accuracy Numbers

Test no Positive Negative Accuracy

Number of images

1 200 185 15 92.5%

2 200 187 13 93.5%

3 300 279 21 93.0%

4 300 282 18 94.0%

5 300 286 14 95.3%

Table 6.2: Testing Accuracy Numbers

CONCLUSION

DEPT.OF CS&E, GMIT, Bharathinagara Page 16

 FACE RECOGNITION ATTENDANCE SYSTEM

The proposed facial recognition-based auto attendance system is many times more efficient than its
fingerprint or RFID based counterparts as no manual intervention is required. Saves time as the
system works for several people simultaneously. The program is relatively lightweight and can be
run on inexpensive hardware. The input device could be integrated to a complete Raspberry Pi
system or could be anything another device like a mobile phone or a webcam.

REFERENCES

DEPT.OF CS&E, GMIT, Bharathinagara Page 17

 FACE RECOGNITION ATTENDANCE SYSTEM

1. honen, Timo, Abdenour Hadid, and Matti Pietikainen. “Face description with local binary
patterns: Application to face recognition.” IEEE transactions on pattern analysis and machine
intelligence 28.12 (2006): 2037– 2041.
2. Ojala, Timo, Matti Pietikainen, and Topi Maenpaa. “Multiresolution gray-scale and rotation
invariant texture classification with local binary patterns.” IEEE Transactions on pattern
analysis and machine intelligence 24.7 (2002): 971–987.
3. Ahonen, Timo, Abdenour Hadid, and Matti Pietikäinen. “Face recognition with local binary
patterns.” Computer vision-eccv 2004 (2004): 469–481.
4. LBPHOpenCV:
https://docs.opencv.org/2.4/modules/contrib/doc/facerec/facerec_tutorial.html#local-binary-
patterns-histograms
Local Binary Patterns:
http://www.scholarpedia.org/article/Local_Binary_Patterns

DEPT.OF CS&E, GMIT, Bharathinagara Page 18