Advanced Liveness Detection in Facial Recognition with Disguises

In recent decades, technology for recognising faces has advanced, making it popular in online communities and complex recognition systems. Face recognition is harder when people conceal themselves. Masked face recognition performs poorly in restricted spaces and big datasets. Face recognition systems must recognise disguised faces for security and dependability. This applies especially when individuals intentionally exploit the system. Ignoring disguised faces as actual faces is a major issue. Masks appear genuine because of their unintentional nature. Disguised faces are changed for style, culture, or personal preference, unlike spoofed ones. These individuals actively participate in recognition, making their masked faces true representations. Masked faces are treated as actual faces, authenticating people by their existence. By considering disguised identities as live faces, aliveness detection algorithms must handle their difficulties. Detecting spoofing became less important than proving identification, even with disguises. New techniques and algorithms are needed to accurately gather and analyse facial features from masked people to determine individual identification. We create an ensemble model to identify and distinguish real and counterfeit faces to address this issue. The ensemble model enhances face recognition by combining RGB and LBP photo texture information. Traditional methods are less accurate and reliable than the ensemble model, which is more resilient.