Biometric systems serve an important function in the identification of individuals, assisting in global security. There are several possible biometrics, such as Genetics, signature, and so on; however, computer vision (CV)-based biometric scanners have gained an important position in the field of individual identity. Customer security is one of the most important components of CV systems. It was proved that basic credentials and login information are ineffective and easily obtained by cybercriminals. CV-based biometrics encompasses the recognition of faces, palmprints, and iris, among other things, and the use of such capabilities to build successful strong authentication. In this study, iris recognition is used, non-contact authentication technologies which are both hygienic and highly reliable. We investigated iris recognition utilizing publicly released databases CASIA Iris Thousand. The obtained iris images are sent through pre-processing procedures. Next, the segmentation and normalization techniques are implemented on processed iris images. Two feature extraction methodologies are applied such as Gray-Level Co-Occurrence Matrix (GLCM), and HAAR to get the key characteristics from normalized images. The two featured data are then utilized to train the Machine Learning (ML) model namely Support Vector Machine (SVM), and K-Nearest Neighbour (KNN). Lastly, the best feature extraction and ML model is identified using the performance measure. The GLCM + SVM delivers a maximum accuracy rate of 97% when compared to other combinations like GLCM + KNN, HAAR + SVM, and HAAR + KNN.