7th International Conference on ICTCS 2022

Authors - Al Mahmud Zayeef, Rana Jyoti Chakma
Abstract - Face recognition technology, defined as the process of detecting and verifying individuals' identities via image processing techniques in computer vision, has become increasingly vital in today's technologically and digitally evolved world. Automating the attendance system with face recognition software eliminates inefficiencies like tampered records, proxies, and security risks. This paper proposes incorporating facial recognition technology, Haar Cascade classifiers, and the Local Binary Pattern Histogram (LBPH). Machine learning approach can be used for identifying faces to create a secure and reliable solution for automating attendance systems. The proposed method produces more accurate results than any other algorithms used earlier. Compared to existing algorithms, the suggested system showed better results, reliability, and robustness in accurately recognizing faces. Its facial recognition accuracy is 87%, and the false positive rate is only 7%. The system can also reliably identify a person who has undergone unintentional changes in their appearance, such as growing a beard or putting on spectacles.