Commentary: Unfolding the role of biometric identification procedures in the current digital era.

Personal identification through different biometry systems is being utilized worldwide on a massive scale. Biometric identification techniques have now penetrated the systems and are being used extensively in hospitals, hotels, public places, etc., This is being used for easy identification of a person, security checks at airports, and marking daily attendance in hospitals and other workplaces.[1] The introduction of the UIDIA-Unique identification authority of India-Aadhar card and its linkage to individual fingerprints and iris scans has given each Indian citizen personal identification and a great sense of security. The different biometric systems can be divided into physiological biometrics or behavioral biometrics. The different physiological biometrics include the following: a) Fingerprints – These are the most popular biometric system. The advantage is that an individual’s fingerprints remain unchanged during a lifetime. b) Hand geometry – this system uses geometric features like finger length and hand width to identify an individual. c) Finger vein pattern – scans of veins are used for individual identity. d) Facial image – this uses facial features like the distance between eyes, nose, mouth, or jaw edges for identification. e) Iris pattern – this allows identification through intricate patterns of the iris, including the furrows and ridges. f) Retinal images – this is considered the most secure way of identification.[2] An individual’s retinal/choroidal pattern is captured and allows for comparison with the database. Various behavioral biometrics include a) Voice verification – this works by recognizing a person’s voice. B) Keystroke identification – this works on the principle of typing signatures, which is based on the pattern of how a person types on the keyboard. C) Handwritten signatures – this helps in authentication based on their handwritten signatures. Biometric evaluation is based on the evaluation of data quality, its usability, and the security measured by correct identification.[3] Hollingsworth et al.[4] studied the effect of pupil dilation on the accuracy of iris biometrics. They found worse recognition performance among highly dilated pupils than constricted pupils. The mean hamming distance also increased with an increase in pupil dilatation size. Dhir et al.[5] in their prospective, non-comparative cohort study of 50 subjects, captured the image before enrollment and 5,10, and 15 min before posting for routine cataract surgery. Two weeks post cataract surgery, the images were again captured. The study’s results were promising as perfect identification and verification were achieved with 0 false rejection, 0 false acceptance, and zero equal error rates. The authors concluded that the reliability of matching decreases with an increase in pupil size. Further, no significant effect of cataract surgery was found on the iris pattern. Seyeddain et al.[6] in another prospective, non-randomized trial studied the reliability of biometric iris recognition for iatrogenic pupil dilation or personal authentication after cataract surgery. Out of 173 eyes, 164 eyes could be identified postoperatively, and the remaining nine eyes were re-enrolled and recognized successfully. In another cohort (group 2), 184 eyes were enrolled in miosis, out of which 22 were not recognized successfully and needed re-enrollment. There was not even a single false positive case. They concluded that iris reorganization is a valid biometric recognition method after cataract surgery or pupil dilatation. Nigam et al.[7] studied the IIT Delhi Cataract Surgery Database from 132 patients employing three commercial iris sensors and assessed the iris texture patterns. The performance was lower when preoperative images were matched to post-operative images (74.69 ± 9.77%) than when preoperative images were compared to preoperative images (93.42 ± 1.76%). In the healthy iris cohort of 68 patients, 100% iris recognition was observed. Their study showed that cataract surgery impacts the iris recognition pattern, and this could be a concern for the reliability of the iris-based biometric recognition system. The current study[8] is a large-scale analysis of the Indian population, and the authors must be congratulated for this interesting analysis. The authors have analyzed the effect of pupil dilation on a biometric iris recognition system for personal authentication and identification by using the iris scanning device “IRITECH-MK2120U” (Biometronic Technology Pvt Ltd, Pyramid North Square, Yelahanka, Bengaluru, Karnataka, India). The iris recognition status and hamming distance were measured for each subject. Matching was performed for all the subjects before and after dilatation. The same person performed all the scans to avoid any interobserver bias. All 321 subjects were matched in the present study, confirming the machine’s reliability. It could be very well concluded that the biometric iris recognition system is a reliable method of identification and authentication in medically dilated normal pupils. This is probably the second study analyzing the iris recognition pattern in the Indian database, the most extensive dataset from India. The authors must be congratulated for this interesting analysis. Besides pupil dilatation, the impact of age, contact lenses, ocular diseases, image compression by devices, the effect of various devices, and various other factors can be studied in the Indian population and is an this is an area of active research in the future.