Face recognition using Skin Color Segment and Modified Binary Particle Swarm Optimization

Face recognition (FR) is a method for identifying or verifying a person's identity based on their face. FR is a research topic that has received a lot of attention, because face identification can be used as biometric security. FR is one of the more popular metric security formats compared with other metric security formats. Face identification also includes a lot of factors that can affect face identification, such as background, head posture, and brightness. The experimental results of the previously proposed methods when encountering color images were not satisfactory. Hence, in this paper, we have presented a method to improve face identification in color images by using three color spaces: RGB, HSV, and YCbCr. The proposed method was tested against the FEI and FERET face databases, and the results were satisfactory compared to other methods where human skin tone was involved in facial identification.