Representation of Hand Dorsal Vein Features Using a Low Dimensional Representation Integrating Cholesky Decomposition

Dorsal hand vein pattern is a promising biometric which is attracting the attention of researchers, of late, to provide more secure identification system. Different approaches have been developed to extract the vein pattern. However, there is a need to find more efficient methods which can reduce matching time. In this work, principle component analysis (PCA), which is a successful method applied on human faces and hand geometry, is being modified based on Cholesky decomposition to represent low dimensional features of the vein pattern. Cholesky decomposition is used to simplify the matrices and it is noticed that there is no loss of information regarding the matching of the eigenveins. The time taken for the processing phase is reduced by 6s which is desirable when developing biometric security system. The system was tested successfully on a database of 200 images with a threshold value of 0.9.