Robust Face Recognition System for Identifying Disguised Individuals

Massive advancements in face recognition over the previous few decades have led to its widespread use in many fields, including social networking and advanced identification systems. The process of recognizing masked faces while accounting for the visual differences caused by factors like camera angle, lighting, and the coverings themselves. Modern techniques for facial recognition are embraced and enhanced in order to appropriately handle disguised faces. The proposed approach focuses on feature extraction employing a wide variety of kernel and filter sizes to detect both local and global features in masked faces. Another strategy to enhance identification performance is the use of skin segmentation algorithms. The purpose of these methods is to isolate the facial region from the masking artifacts. Comprehensive experiments are conducted using benchmark datasets to assess the efficacy of the planned technique. Using assessment system of measurement including recall, accurateness, and exactness, we test the models' ability to identify disguised faces. These measures are used to assess the approaches. According to the results, the ensemble model is more reliable and accurate than competing methods, and it also has better robustness.