Personal recognition using human eyes is no more a science fiction. It has been successfully deployed and being used in several application across the world. Many present day smartphones leverage iris recognition and can unlock the device or authenticate a transaction just by scanning eyes with on-board camera. Internal structure of a human eye offers a number of opportunities for personal identification. Iris and Retina, which are found inside the human eye, feature patterns that can be measured with the help of technology. On retina, this pattern is made by the blood vessels responsible for retina’s blood supply, while on iris, folding of muscular ring creates a random pattern. Both retinal and iris patterns vary greatly from person to person and can be used to uniquely identify them. Secure behind the transparent cornea, iris is visible with naked eye from outside, while retina, being on posterior portion in the eye, can only be viewed with special equipments, called retinal scanners.
Human eyes: more distinctive than you know
While fingerprint was the most used modality in biometric applications in the beginning, other human anatomical and behavioral characteristics were also taken into account later. It paved the way to several other human traits to be used for technology powered personal identification including eyes. A lot has been said and written about human eyes. Eyes provide us with vision. Several parts of this organ works together to help us percept the visible world. They help us recognize things from a distance, but with biometrics, eyes can help us get recognized as well. Any human behavioral or physiological trait has to have some set attributes to be a biometric characteristic.
Universality: It should be present in each subject of target population.
Uniqueness: It should be unique for each individual. For example, facial features are unique but some individuals and twins may have their facials features repeating.
Permanence: It should be stable over a period of time. Patterns that do not change during the lifetime of an individual are more suitable for the purpose.
Collectability: The process of acquisition of characters should be user friendly. A biometric characteristic should be collectable with or without special means.
Performance: It should be good enough for a biometric system to achieve the desired accuracy for which it was designed.
Acceptability: A biometric identifier is expected to be accepted by the target population. For example, fingerprint recognition is more acceptable than retina or DNA sampling. Invasiveness of sampling process of a biometric identifier can be a deciding factor for its acceptability. Sampling of fingerprints requires just a touch while sampling of DNA requires physical collection of sample.
Fortunately, retina and iris in human eyes fulfil all above requirements to be leveraged as a biometric identifier. There are specialized systems that have been already deployed and are being constantly improved for iris and retina recognition. Both the methods have their own set of advantages and disadvantages. Let’s dig little deeper into both the methods, one by one.
The posterior portion of human eye forms retina. It is made of a light sensitive tissue. When light passing through cornea and lens reaches retina, neural signals are generated and transferred to the brain via the optic nerve. Retina is a thin layer of tissue formed by neural cells. Capillaries responsible for blood supply of this layer forms a pattern that can be used for personal identification. This pattern of blood capillaries is believed to be unique in each individual due to huge possibility of variation how these capillaries run on the surface of retina. Since retina is located at the posterior portion inside the human eye, special equipment is required to scan this pattern. Retina recognition is one of the least deployed biometric methods because of high cost of the implementation and its highly invasive nature that may cause some user discomfort. Still, it is used is very high security applications like military and high level government access due to its accuracy and high level of security.
Retina recognition systems make use of low energy infra-red light to scan the retinal pattern. Blood vessels absorb infrared light while surrounding tissues reflect it. This reflection is detected by the retina recognition system and image of this pattern is captured. This image is further enhanced to make is usable for the recognition algorithm. Retina template is generated once the image is taken through recognition algorithm; this template is associate with a subject’s demographic data and stored. The process so far is called enrolment. The subject’s identity can be verified anytime by scanning a new retinal sample and matching it against the stored template.
Iris is the ring shaped colored portion in a human eye and is visible from outside with naked eye. It is made of muscle tissue that adjusts the size of pupil and controls how much light can enter the eye. Amount of melatonin pigment in iris is responsible for different colors that human eyes take. Folds in iris muscles throughout the ring create a pattern with great amount of details. Formation of this pattern is completely random and there is no rule how it will turn out in an individual’s eye. However, once this pattern is created during the foetal development, it stays the same throughout the life. An individual’s irises are unique and structurally distinct, even iris of same individual does not match. All these attributes make them good enough for personal recognition.
Details of iris can be captured with any high quality digital camera, however, modern recognition systems make use of near infrared (NIR: 700–900 nm) instead of visible light to capture details. Since iris recognition can be established with high quality camera and recognition software, it can be setup on any computing device; however, dedicated recognition systems are more common due to performance and security reasons. Iris recognition systems use a camera to capture details of the iris and this image is enhanced by the image enhancement algorithms. Once the image is usable enough, it is processed by the recognition algorithms, which extracts unique features to generate a biometric template. Associating identity data with this template establishes identity of the subject in question, which can be used for identity verification in future.
This table compares retina and iris recognition side by side:
|Non-invasive, modern iris recognition systems can scan from a distance, user just need to look at the imaging equipment||Highly Invasive. A light beam is concentred on the subject’s eye. Subject needs stare at equipment optics without moving during the sample collection|
|Visible from outside||Can only be scanned by specialized equipment|
|User consent is required, however some applications may not need this||User consent is required|
|Used for personal as well as large scale public applications like airports||Used in personal identification and high security application|
|Subjects can be identified from a distance||Subjects cannot be identified from a distance|
|Reasonably secure and widely deployed||Highly secure but not widely deployed|
|High permanence and stability, may get affected by certain diseases and age||High permanence and stability, may get affected by certain diseases and age|
|Highly Collectable, details are visible from outside||Low collectability, details are inside the eye.|
|Comparatively simpler setup is required than retina recognition||Specialized setup is required to collect and process sample|
|High universality||High universality|
|High potential for circumvention||Low potential for circumvention|
|High acceptability||Low acceptability|
Real life deployments of retinal and iris recognition systems
Biometric recognition using human eye is getting momentum in commercial as well civil identification needs. Iris is found to be more appropriate modality that offers a balance in cost of implementation, security and user convenience. It is increasingly outnumbering retina based recognition systems.
Travel and border control (iris)
United Arab Emirates’ border control agencies have been using iris recognition at all border security checkpoints on land, air and sea ports. Travellers with visitor visa, who are not the citizen of the country, have to go through iris recognition process to enter the UAE. Similar efforts are being made in Canada, which is using iris recognition technology under CANPASS Air program, which is used to speed through low risk pre-approved travellers. Currently the system is operational in several Canadian Airports.
Civil identification (iris)
Iris recognition application for civil identification has been deployed for government and security agencies throughout the world. However, India’s Aadhar program is the most notable example of iris recognition application. It hold iris and fingerprint templates of the more than 1.2 billion Indian citizens, which is considered to be the world biggest biometric database. Similar efforts have been made
Mobile biometrics (iris)
History of biometric recognition on mobile phones dates back to 2007, when Toshiba launched its Windows Mobile devices with fingerprint sensor. However, iris scanner did not make it to mobile phones until 2015, when Fujitsu launched the first smartphone with iris scanning capabilities. After that technology firms has launched many mobile devices with iris recognition, Samsung Galaxy Note 7, S8, Note 8, Microsoft 950 XL, Vivo X5 Pro are among the most notable ones. Unlike iris recognition, which can be setup using a digital camera and recognition software to process data, retina recognition require special setup to illuminate retina and capture the pattern of blood vessels, which cannot be setup in today’s space constraint devices. On the other hand, iris recognition offers a viable solution and is already being deployed by mobile phone manufacturers.
Government agencies (retina)
Retinal scanning has been used by government agencies like CIA, FBI, NASA, etc. It has also been used in for customer identification at ATMs and prisoner identity verification at correctional facilities.
Challenges in iris and retinal recognition
Despite the rapid advancement in technology, there is no biometric recognition system that can be called perfect. Recognition systems still have many shortcomings related with security and convenience that need to be addressed. Users tend to like least invasive techniques that also let them know that they are being identified. It is the probable cause of success of Fingerprint Identification as it keeps a balance of convenience, invasiveness and security. Hands can be moved freely and fingertips can be touched the way a user wants, however that is not the case with eye based identification methods. They cause a certain degree of user inconvenience and make them conscious, specially in case of retina recognition. The user has to keep his/her eyes steady, especially in case of retinal scan.
Despite these shortcomings, retina scan is a highly secure method of identification and immune to spoof attacks. Iris scan, however, is more practical in real world applications. It is more prone to spoof attacks than retina recognition. Other than challenges specific to iris or retina recognition, usual risk factors with biometric recognition, biometric data security and performance of biometric systems are the common challenges for both the recognition methods.
Eyes play an important role in linking us with visible world and precept it. Patterns formed by blood vessels in retina and muscles in iris can help identify individuals. Being inside the eyes, both the recognition methods are considered to be secure, however, retinal recognition is used in very high security access as it is not visible from outside without special equipments. Iris and retina patterns have been found to be unique to an individual due to their complexity and enormous possibility of variations. Some biometric software solution providers also capture details of sclera (the white of the eye with visible thin blood vessels) to establish identity. This extra detail is leveraged to control deviations occurred due to user behavior.