Vein mapping can be used to identify possible suspects using matching learning algorithms. Since vasculature deep in the skin cannot be visualized by naked eyes, the features extracted usually by converting to near infrared images which gives best track recovery with little noise. Two decades, ago the premise for the use of vein patterns for identification emerged in the forensic field. Researchers are proposing innovative approaches and methods utilized to improve the recognition, quality, classification, and extraction of viable vein patterns from images. Deep learning algorithms such as convolution neural network (CNN ), K-nearest network, autoencoders are being used to extract venous features with ease especially when analyzing image forensic evidence. This paper provides an overview of recently proposed finger vein, dorsal hand vein, wrist vein and hybrid systems and highlights their performance and real-life application.
