VISION-BASED AUTOMATED ATTENDANCE SYSTEM USING HAAR CASCADE ALGORITHM

Maintaining accurate student attendance records in educational institutions has become

increasingly challenging, particularly with the rapid expansion of virtual and hybrid learning environ-
ments. Traditional attendance systems that rely on manual recording are time-consuming, inefficient,

and highly susceptible to human error. Moreover, such systems are vulnerable to fraudulent practices,
including proxy attendance, which compromises the reliability and integrity of attendance data. These
limitations highlight the need for an automated and intelligent attendance monitoring system that

ensures accuracy, efficiency, and transparency. Facial recognition technology has emerged as a reli-
able biometric identification method due to the uniqueness and difficulty of replicating human facial

features. This paper proposes a real-time automated attendance system that utilizes face detection and
recognition techniques to accurately identify students. The system captures live video input through
a camera and processes the frames to detect and recognize faces. Detected faces are compared with
stored facial data in a database, and attendance is automatically recorded when a successful match is
identified. The proposed system employs the Haar Cascade algorithm, a machine learning–based
object detection technique widely used for real-time face detection. The algorithm is trained using

a large dataset of positive (face) and negative (non-face) images, enabling efficient and accurate de-
tection of frontal faces. The system is implemented using OpenCV, an open-source computer vision

library that provides robust tools for image processing and facial recognition applications.