Detecting Fake Face Images in Biometric Systems Using LBP Feature Extraction and CNN

Face spoofing has become an increasing concern in terms of security. It involves using misleading face images or videos to trick authentication systems. This study uses convolutional neural network (CNN) and local binary patterns (LBP) approaches for the extraction of features to propose a reliable solution for detecting face spoofing. It utilizes LBP to extract unique features from facial pictures. A CNN model that has undergone substantial training with a sizable dataset is then fed these extracted features. The purpose of this training is to distinguish between genuine and fake faces. CNN’s ability to automatically learn advanced features enables it to effectively identify subtle differences between real and counterfeit faces. Moreover, the implications of this research extend to various domains such as mobile banking, access control, and surveillance. This development marks an important milestone in safeguarding against deceptive practices and reinforcing the trustworthiness of face-based authentication mechanisms.