Abstract: Retina authentication technology’s captures and analyzes them patterns off vascular on them thins nerve on the backs of them eyeball that processes lights entering through the pupil. Retinal patterns are highly distinctive traits; It cannot begotten or stolen. No, two people have same retinal pattern and the features extracted from retinal image are the most stable and reliable than other biometrics features. Due to these reasons the retina is chosen as robust source in the proposed authentication systems. In this thesis, a system for recognition based on vascular pattern of human retina is introduced. This system contains three stages (preprocessing, features extracting and matching). Preprocessing stage is used in order to make retina image more clear by using Gaussian and Canny filters and Haar wavelet transformer. Features extracting phase is done using two types of inspired algorithms (Bat and Harmony). Each extracted feature list is assembled as a feature vector which is used in order to distinguish between different retinal images. Finally, the discrimination analysis of features is evaluated in matching phase by using Euclidean distance measure. The results from this research declared that the best inspired algorithms for the proposed retinal authentication system are harmony algorithm in which the obtained accuracy is (100%) while the accuracy of bat algorithm (98.65%). However, harmony algorithm is the best for the proposed authentication system in time consuming which is (3.0 sec) and bat consuming time is (7.3 sec).