2D/3D spatial distribution of superficial blood vessels in human skin in vivo was conducted by double correlation analysis of the swept source Optical Coherence Tomography (OCT) images. An adaptive Wiener filtering technique has been employed to remove background noise and increase the overall quality of the OCT images acquired experimentally. Correlation Mapping and Fourier domain correlation approaches have been subsequently applied to enhance spatial resolution of images of vascular network in human skin. The analysis of images performed on Graphics Processing Units (GPUs) utilizing the recently developed Compute Unified Device Architecture (CUDA) framework.