Entropy of the Retina Template

Abstract

Abstract. We compare two vessel extraction methods for creation of a retina template, using a database of 20 images of normal retinas. Each vessel in a well defined region is represented by a three dimensional feature, from which a retina template is built. Based on the sample distributions, we propose a preliminary theoretical model to predict the entropy of a retina template. We analyse by experimental and theoretical means the entropy present, and infer that entropy from our retina template compares sufficiently favourably with that of a minutia-based fingerprint template to warrant further study.

A New Retinal Recognition System Using a Logarithmic Spiral Sampling Grid

“…We carried out the evaluation of the recogniton system using separately each time series as the feature descriptor, i.e., the time series of the position and that of the crossing angle. It was concluded from the experimental results that we obtain lower error rates using as the feature vector the position time series; this result is consistent with that obtained by [27] concluding that the position of the blood vessel as a descriptor is the main contribution to retina template entropy. If the FAR and FRR rates for a series of threshold values are obtained, it is possible to plot these rates as a Receiver Operating Curve (ROC), where one axis displays FAR rates and the other displays FRR rates.…”

State of the Art in Vascular Biometrics

“…The actual information content of retinal data has been investigated in some detail [232], with particular focus set on minutiae-type [103,232] and vesselrepresentation-type templates [7], respectively.…”

Retina features based on vessel graph substructures

“…More recent support for the distinctiveness of retina tem plates than the original sources [14,15], appears in [6], where the entropy of their vessel-crossing template is es timated as superior to that of a minutia-based fingerprint template, and in [] 1], where the entropy of the feature point based retina template is estimated as comparable to that of iris [16].…”

“…These include extraction of restricted patterns of vessels crossing the retina boundary [4], or traversing a circular region centred on the fovea [1] or an annulus centred on the optical disc [5]. Vessel cali bre was added in [6] as an additional feature to the template described in [5]. Templates based on feature points (posi tions of vessel branchings and crossovers) have been tested recently in [7,8,9,10,11,12] and the effect of matching 1 such features in fovea-centred versus optical disc-centred images has been studied in [13].…”