IRIENCE - High security solution group

Strength of IRIS Recognition

  • Proven highest accuracy: iris recognition had no false matches in over two million cross-comparisons
  • Ability to handle very large populations at high speed: can handle very large 1: all searches within extremely large databases.
  • Convenient: all a person needs to do is look into a camera for a few seconds. A video image is taken which is non-invasive and inherently safe.
  • The iris itself is stable throughout a person’s life (approximately from the age of one); the physical characteristics of the iris don’t change with age.
  • Total Cost of Ownership: iris recognition carries extremely low maintenance costs and offers seamless interoperability between different hardware vendors; the technology also has the ability to work well with other applications.

During the development of the iris, a process known as “chaotic morphogenesis” occurs during the seventh month of gestation, which means that even identical twins have differing irises.
Iris is protected behind the eyelid, cornea and aqueous humour means that, unlike other biometrics such as fingerprints, the likelihood of damage and/or abrasion is minimal. The iris is also not subject to the effects of aging which means it remains in a stable form from about the age of one until death. The use of glasses or contact lenses (coloured or clear) has little effect on the representation of the iris and hence does not interfere with the recognition technology.

Accurate Every iris is absolutely unique. None of other biometric recognition technology can lower the FAR(False Accept Rate) below 0.0001% which iris recognition technology achieve very easily. There is no argument about the fact that iris recognition technology is the best and most accurate among biometrics. Iris sdk
Fast Looking at speed in conjunction with accuracy, there’s simply no other technology that can deliver high accuracy authentication in anything close to the real-time performance of iris recognition.
Scalable As iris data templates require only 579 bytes of storage per iris, very large databases can be managed and speedily searched without degradation of performance or accuracy.
Non-invasive Iris recognition is non-contact and quick, offering unmatched accuracy when compared to any other security alternative, from distances as far as 10 to 60 cm.

Iris Recognition Process

  1. Capturing the image
    The image of the iris is be captured using a Irience camera using near infrared light. The camera is positioned between 31 and 34cm and to capture the image. In the manual procedure, the user needs to adjust the camera to get the iris in focus and needs to be within this range of the camera. This process requires initial user training to be successful.
  2. Defining the location of the iris and optimizing the image
    Once the camera has located the eye, Irience system identifies at 30 frames/sec the image that has the best focus and clarity of the iris. The image is then analysed to identify the outer boundary of the iris where it meets the white sclera of the eye, the pupillary boundary and the centre of the pupil. This results in the precise location of the circular iris.
    Our iris recognition system then identifies the areas of the iris image that are suitable for feature extraction and analysis. This involves removing areas that are covered by the eyelids, any deep shadows and reflective areas.
  3. Storing and comparing the image.
    Once the image has been captured, the algorithm filter and map segments of the iris into hundreds of vectors Algorithms also take into account the changes that can occur with an iris, for example the pupil’s expansion and contraction in response to light will stretch and skew the iris. This information is used to produce what is known as the template, which is a 579-byte record. This record is then stored in a database for future comparison. When a comparison is required the same process is followed but instead of storing the record it is compared to all the template records stored in the database. The comparison also doesn’t actually compare the image of the iris but rather compares the hexadecimal value produced after the algorithms have been applied. In order to compare the stored template record with an image just scanned, a calculation of the Hamming Distance is required. The Hamming Distance is a measure of the variation between the template record for the current iris and the template records stored in the database i.e. bit 1 from the current template and bit 1 from the stored template record are compared, then bit 2 and so on. Any bits that don’t match are assigned a value of one and bits that do match a value of zero. Once all the bits have been compared, the number of non-matching bits is divided by the total number of bits to produce a two-digit figure of how the two template records differ. For example a Hamming Distance of 0.20 means that the two template differ by 20%.

With all biometric systems there are two error rates that need to be taken into consideration. False Reject Rate (FRR) occurs when the biometric measurement taken from the live subject fails to match the template stored in the biometric system. False Accept Rate (FAR) occurs when the measurement taken from the live subject is so close to another subject’s template that a correct match will be declared by mistake. The point at which the FRR and the FAR are equal is known as the Crossover Error Rate (CER). The lower the CER, the more reliable and accurate the system. In iris recognition technology, a Hamming Distance of .342 is the nominal CER. This means that if the difference between a presented template record and one in the database is 34.2% or greater then they are considered to have come from two different subjects. During recognition mode, this comparison has to occur between the template record from the live subject and every template stored in the database before the live subject is rejected. The following table shows the probabilities of false accept and false reject with iris recognition

False Accept
False Reject
0.28 1 in 1012 1 in 11,400
0.29 1 in 1011 1 in 22,700
0.30 1 in 6.2 billion 1 in 46,000
0.31 1 in 665 million 1 in 95,000
0.32 1 in 81 million 1 in 201,000
0.33 1 in 11 million 1 in 433,000
0.34 1 in 1.7 million 1 in 950,000
0.342 1 in 1.2 million 1 in 1.2 million
0.35 1 in 295,000 1 in 2.12 million
0.36 1 in 57,000 1 in 4.84 million
0.37  1 in 12,300 1 in 11.3 million


IRIENCE SDK, defined as; Monro Iris Recognition Library and Interface is the name given to Smart Sensors’ fully featured toolkit which enables our partners to build iris recognition engines, enrolment and ID management applications that use iris image or template databases.

Features : excellent cross-platform support: Windows, Linux, Embedded, DSP

Iris/ Pupil Finder
very rapid location of iris and pupil
ideal for camera developers

Liveness Detection
IRIENCE includes functions that enable camera and systems developers to include liveness and spoof detection functions by analysing real-time changes in pupil size and templates produced from the iris texture. Watch a video showing how IRIENCE can track real-time pupil and iris changes.


In September 2009, NIST (National Institute of Standards and Technology, Gaithersburg, MD, USA) reported on testing conducted under the IREX I project in support of the development of a new version of ISO/IEC 19794-6 standard for iris images intended for use in biometric applications. This was released during 2011.

The original parent image, and 4 derivatives were tested – see below:

Please note that the above diagram is taken from the original draft ISO document in which the term ‘kind’ was used. This has, as per ISO 19794-6:2011, been replaced with ‘type’. As much of the information in this section is derived from the draft document, we use the term ‘kind’, but the user should be aware that this is no longer the preferred usage.

KIND 2 is a VGA simple cropped image with the pupil in the centre (not shown in above diagram).

KIND 3 is a simple cropped image with a margin so that the horizontal dimension is about 160% of the nominal iris radius and the vertical dimension about 120% of the nominal iris radius.

KIND 7 is a fully segmented image, see table below.

KIND 16 and KIND 48 were rejected (see later text)

An illustration of an ROC for a typical set of images using the MIRLIN algorithm is presented below. The crossover point represents the point at which the False Reject Rate is equal to the False Accept Rate, and is called the Equal Error rate or EER.