Individual recognition through palm vein authentication has gained the attention of the scientific community due to its high level of security. However, the algorithms for recognition are validated with a limited number of images due to the small number of subjects in public databases, making it challenging to implement deep learning-based methods and evaluate scalability for mass identification. Creating a large-scale database of real palm vein images is laborious in terms of time, security, and cost. In other biometrics, such as fingerprint recognition, synthetic images greatly enhance the accuracy of developed techniques. Although the reasons behind palm vein patterns are not fully understood, there is evidence that geometric characterization and anatomical study allow for the proposal of reasonable assumptions to create realistic vein pattern images through models. Therefore, this study aims to generate synthetic palm vein images by modeling the vascular structure. Thus, our proposal will favor future research that requires the generation of large-scale databases to provide reliable and scalable solutions to biometric recognition tasks of individuals with palm veins.