Adoptability of Open-Source Face Recognition (FR) on Automated Criminal Identification System for Law Enforcement in the Philippines: A Systematic Review
William Penaflor Rey, Eric Blancaflor, Kieth Wilhelm Jan D. Rey · 2022
Crime prevention in society has persisted in being a significant issue that needs to be controlled through efficient investigation processes. Among these methods, biometric facial recognition is particularly effective because it is adaptable and non-intrusive. Face recognition has grown in popularity due to its numerous uses in entertainment, smart cards, information security, law enforcement, and surveillance. It is a vital topic in pattern recognition, computer vision, and image processing. For feature extraction, two basic methods are used: appearance-based and model-based. Methods based on appearance employ global Face identification using representations. Model-based face approaches seek to create a human face model that captures facial diversity. The distance between the vectors of two photographs is defined as image similarity. This research primarily analyzed open-source Face Recognition algorithms based on the Labeled Faces in the Wild (LFW) model. The study will be of significant consideration in integrating Face Recognition technology on the current Mobile Automated Fingerprint Identification System (MAFIS) using Minutiae Structure.