[Paper Review] Automating Attendance Management in Human Resources: A Design Science Approach Using Computer Vision and Facial Recognition

Tat-Bao-Thien Nguyen, Minh-Quoc Bui+1 more|arXiv (Cornell University)|May 21, 2024

COVID-19 Pandemic Impacts4 citations

TL;DR

This paper proposes a low-cost, embedded facial recognition system for automating HR attendance management using Haar Cascade and OpenCV2 on an NVIDIA Jetson Nano. By leveraging efficient image processing and edge detection, it achieves accurate face detection and matching with minimal manual intervention, offering a scalable, affordable solution for schools, small businesses, and workplaces with limited computational resources.

ABSTRACT

Haar Cascade is a cost-effective and user-friendly machine learning-based algorithm for detecting objects in images and videos. Unlike Deep Learning algorithms, which typically require significant resources and expensive computing costs, it uses simple image processing techniques like edge detection and Haar features that are easy to comprehend and implement. By combining Haar Cascade with OpenCV2 on an embedded computer like the NVIDIA Jetson Nano, this system can accurately detect and match faces in a database for attendance tracking. This system aims to achieve several specific objectives that set it apart from existing solutions. It leverages Haar Cascade, enriched with carefully selected Haar features, such as Haar-like wavelets, and employs advanced edge detection techniques. These techniques enable precise face detection and matching in both images and videos, contributing to high accuracy and robust performance. By doing so, it minimizes manual intervention and reduces errors, thereby strengthening accountability. Additionally, the integration of OpenCV2 and the NVIDIA Jetson Nano optimizes processing efficiency, making it suitable for resource-constrained environments. This system caters to a diverse range of educational institutions, including schools, colleges, vocational training centers, and various workplace settings such as small businesses, offices, and factories. ... The system's affordability and efficiency democratize attendance management technology, making it accessible to a broader audience. Consequently, it has the potential to transform attendance tracking and management practices, ultimately leading to heightened productivity and accountability. In conclusion, this system represents a groundbreaking approach to attendance tracking and management...

Motivation & Objective

• To address the high cost and complexity of existing deep learning-based attendance systems in HR and educational settings.
• To develop a resource-efficient, accurate, and scalable attendance management solution suitable for small to medium institutions.
• To reduce manual data entry and human error in attendance tracking through automated facial recognition.
• To demonstrate the feasibility of using Haar Cascade with OpenCV2 on embedded hardware for real-time face detection and matching.
• To provide an affordable, accessible alternative to commercial facial recognition systems for institutions with limited computing budgets.

Proposed method

• The system uses Haar Cascade, a machine learning-based object detection algorithm, to detect faces in images and videos using Haar-like features and edge detection techniques.
• It integrates OpenCV2 for image preprocessing, feature extraction, and real-time face detection and matching against a stored database.
• The system runs on the NVIDIA Jetson Nano, an embedded computing platform optimized for low-power, on-device inference.
• Haar features are carefully selected, including Haar-like wavelets, to improve detection accuracy and robustness in varying lighting and pose conditions.
• Face matching is performed by comparing detected facial features with stored templates in a local database using similarity metrics.
• The design emphasizes computational efficiency and low hardware requirements to enable deployment in resource-constrained environments.

Experimental results

Research questions

• RQ1 Can a Haar Cascade-based system achieve high accuracy in face detection and matching for attendance management with minimal computational overhead?
• RQ2 How does the integration of OpenCV2 and the NVIDIA Jetson Nano impact processing efficiency and real-time performance in embedded deployment?
• RQ3 To what extent can a Haar Cascade system reduce manual intervention and human error in HR attendance tracking compared to traditional methods?
• RQ4 How does the system perform across diverse environments such as classrooms, offices, and factories with varying lighting and background conditions?
• RQ5 What is the trade-off between accuracy and resource efficiency when using Haar Cascade versus deep learning models in attendance systems?

Key findings

• The system achieves high accuracy in face detection and matching using only Haar features and edge detection, without requiring large datasets or expensive hardware.
• Processing efficiency is significantly enhanced by deploying the system on the NVIDIA Jetson Nano, enabling real-time operation on embedded devices.
• The use of Haar Cascade reduces computational costs and infrastructure requirements, making the solution affordable and accessible for small institutions.
• The system demonstrates robust performance across diverse environments, including schools, vocational centers, and small offices, with minimal manual configuration.
• By minimizing manual data entry, the system reduces human error and strengthens accountability in attendance tracking processes.
• The solution is validated as a scalable, low-cost alternative to commercial facial recognition systems, suitable for widespread deployment in educational and small business settings.