Face Recognition Based Attendance System using Haar Cascade and Local Binary Pattern Histogram Algorithm

Abstract

Facial recognition attendance systems have garnered considerable attention due to their capability to automate attendance tracking while addressing the limitations of conventional manual methods. These systems employ advanced algorithms, such as Haar Cascade and Local Binary Pattern Histogram (LBPH), to analyse and match facial patterns, enabling precise identification and verification of individuals. This research provides an in-depth investigation into the application of the Haar Cascade and LBPH algorithms within a facial recognition attendance system. The study demonstrates the algorithms' proficiency in accurately recognizing faces, displaying individuals' names, and reliably recording attendance with an impressive accuracy rate of 99.0%. Functionally, the technology captures images or videos of individuals' faces upon their arrival and subsequently compares them to a pre-existing database. Significantly, as the dataset size expands, the system's accuracy exhibits consistent improvement. Notably, the research identifies a threshold for the minimum number of images required to achieve dependable attendance prediction. The results produced indicate the effectiveness of the LBPH and Haar Cascade algorithms in automating attendance tracking, reducing errors, and reducing administrative burden. The adoption of facial recognition attendance systems represents a scholarly and robust solution with broad applicability, ensuring precise attendance records across diverse contexts.