Multibiometric Authentication System Using Gesture Recognition

To successfully increase the security of the system in today's environment, person authentication is required. It is undeniably true that most individuals would prefer to be authenticated in the simplest and most transparent way possible, without having to remember a personal identification number. Identity verification is necessary for many everyday activities. Both the ear and the palm's features are recorded, and corresponding feature vectors are created. The score level values are obtained following the fusion process, and the values below the threshold values are deleted and the values over the threshold values are authenticated using the Gaussian Log Gabor filtering method. In this regard, a multi-biometric system based on a person's gesture can be employed. It uses the complementing physical and behavioral characteristics of the ear and palm, two separate biometrics. The transfer function is the log Gabor method. The system's reliability is effectively increased by the aforementioned technique. The main benefit was created based on the merging of two biometric systems to enhance performance as a whole. Various feature extraction, feature matching, and data fusion approaches are used to determine the saliency and correlation of the data that each sensor has collected. The outcomes also imply that the proposed algorithms outperform existing algorithms in terms of performance.