You are on page 1of 11

FACE RECOGNITION BASED ATTENDANCE

SYSTEM

SUBMITTED BY:
Priyansh Lunawat (221030169)
Vaibhav Raj (221030155)
Aims and Objectives:
The objective of this project is to develop face recognition attendance system.
● To detect the face segment from the video frame.
● To extract the useful features from the face detected.
● To classify the features in order to recognize the face detected.
● To record the attendance of the identified individual.
Scope of the project:
• Attendance Tracking:
o The primary purpose of the system is to automate attendance tracking. It
can be used in educational institutions, corporate offices, and other
organizations where attendance needs to be monitored.
• Accuracy and Efficiency:
o Haar cascade-based face detection is relatively fast and can work in real-
time, making it suitable for applications where quick and accurate
attendance tracking is required.
• Security Enhancement:
o The face detection system can contribute to the overall security of a
facility by ensuring that only authorized individuals gain access. It can be
integrated with access control systems.
• User Identification:
o Beyond attendance, the system can be extended to identify individuals.
This can be useful in scenarios where personalization or individual
tracking is necessary.
• Remote Attendance:
o With appropriate modifications, the system could support remote
attendance tracking. This is especially relevant in the context of remote
work and online education.
• Data Analysis and Reporting:
o The system can generate reports and analytics on attendance patterns.
This information can be valuable for administrators, teachers, or
managers to identify trends and make data-driven decisions.
• Open Source Libraries:
o Python offers several open-source libraries for image processing and
computer vision, making it easier to implement and customize the face
detection system. OpenCV, for example, provides Haar cascades and tools
for image processing.
• Adaptability to Different Environments:
o Haar cascade-based face detection can be adapted to different
environments and lighting conditions. This adaptability makes it suitable
for deployment in a variety of settings.
• Educational and Research Applications:
o The implementation of a face detection attendance system can serve as an
educational project or be used for research purposes in the fields of
computer vision and machine learning.
FLOWCHART
Face Detection Method :
Haar Cascade Algorithm

The Haar cascade algorithm is a machine learning-based object detection


method used to identify objects or features in digital images or video. It was
introduced by Viola and Jones in their 2001 paper "Rapid Object Detection
using a Boosted Cascade of Simple Features." The algorithm is particularly
known for its efficiency in real-time object detection. The Haar cascade
algorithm is widely used for face detection due to its speed and effectiveness.
Haar cascade is an algorithm that can detect objects in images,
irrespective of their scale in image and location. This algorithm is not so
complex and can run in real-time. We can train a haar-cascade detector to
detect various objects like cars, bikes, buildings, fruits, etc.
PROS AND CONS OF HAAR CASCADE ALGORITHM

PROS:
• Fast Real-time Processing:
o Haar cascade-based object detection is known for its speed, making it
suitable for real-time applications. The algorithm is efficient in processing
video streams and images, making it a popular choice for applications
where speed is crucial.
• Low Computational Requirements:
o Compared to some other object detection algorithms, Haar cascades have
relatively low computational requirements. This makes them feasible for
deployment on devices with limited computational resources.
• Adaptability to Different Environments:
o Haar cascades can perform well in various environments and lighting
conditions. The algorithm is capable of handling variations in pose, scale,
and illumination, which makes it versatile for different applications.
• Simple Features:
o The Haar features used in the algorithm are simple and computationally
inexpensive to compute. This simplicity contributes to the speed of the
algorithm.
• Training Efficiency:
o The AdaBoost training process used in Haar cascades is efficient and
helps in creating a strong classifier by combining multiple weak
classifiers. The training process can be completed relatively quickly
compared to some other machine learning-based object detection
methods.
CONS:
• Limited Accuracy for Complex Objects:
o Haar cascades may struggle with detecting complex objects or objects
with intricate patterns. They are particularly effective for objects with
distinctive features but may face challenges with objects that have
complex shapes or textures.
• Sensitivity to Occlusion:
o The algorithm may be sensitive to occlusion, where the object of interest
is partially obscured. This can result in false negatives or incomplete
detections in scenarios where the object is not fully visible.
• Dependency on Training Data Quality:
o The performance of Haar cascades heavily depends on the quality and
representativeness of the training data. If the training dataset is not
diverse enough or does not adequately capture variations in the real-
world scenarios, the algorithm's performance may be compromised.
• Not as Robust to Variations as Deep Learning Approaches:
o While Haar cascades are versatile, they may not be as robust as deep
learning approaches, such as Convolutional Neural Networks (CNNs),
when it comes to handling a wide range of object variations, including
changes in viewpoint, scale, and lighting.
• Manual Feature Engineering:
o Haar cascades rely on manually crafted features (Haar features), which
may limit their ability to adapt to complex patterns. In contrast, deep
learning approaches automatically learn features from the data, allowing
for more flexibility in handling diverse patterns.
Functioning of the project:

This the user interface, it consists of two sections ,


1) New Registration: Takes the individuals Name and ID,then the image of
the individual and saves its profile.
2) For already registered users: Scans the face of the individual and marks
the attendance along with the date and time.

HELP BOX
ATTENDANCE RECORD (EXCEL SHEET)
LIBRARIES USED

Limitations of the project:


• Detection process is slow and high time complexity.
• Long Training Time.
• Limited Head Pose.
• Object must be closer to the camera.
• Can be detected with pictures and Id’s also.
• Sometimes creates problem in detection due to dark
background.
• Face movement should be still or slow.
References:
• https://www.analyticsvidhya.com/blog/2021/11/build-face-
recognition-attendance-system-using-python/
• https://github.com/Patelrahul4884/Attendance-Management-
system-using-face-recognition
• https://chat.openai.com/
• Project report by ANITS

You might also like