학술논문

Facial Recognition System with LBPH Algorithm: Implementation in Python for Machine Learning
Document Type
Conference
Source
2024 Second International Conference on Intelligent Cyber Physical Systems and Internet of Things (ICoICI) Intelligent Cyber Physical Systems and Internet of Things (ICoICI), 2024 Second International Conference on. :1681-1686 Aug, 2024
Subject
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Histograms
Machine learning algorithms
Accuracy
Face recognition
Noise
Lighting
Machine learning
Feature extraction
Real-time systems
Security
Face Recognition
Local Binary Pattern Algorithm
Machine Learning
Haar cascade classifier
Deep Learning
Language
Abstract
Face recognition technology, while making significant advancements, faces challenges in achieving human-level accuracy due to factors like variations in facial appearance, lighting conditions, and noise. This research proposes a novel approach using the Local Binary Pattern (LBP) algorithm to improve face recognition accuracy. By utilizing LBP histograms, our method enables face recognition from both frontal and side views, enhances performance in low-light conditions, and increases real-time identification rates. The primary objective of this research is to develop a robust face recognition system for security applications. The proposed system offers user benefits by preventing unauthorized access and ensuring security. The LBP histogram method is employed for face recognition, requiring users to enter their name and photo into a database. Upon login, the system identifies the user, displaying their name and triggering a warning alert for unknown individuals. The implementation of this face recognition system using the LBP algorithm aims to provide a widely applicable and valuable solution for various organizational settings.