This work is licensed with a
Creative Commons Attribution 4.0 International LicenseEndFragment
This material was developed with funding
from the National Science Foundation
Biometrics
Restart
Next
Back
Introduction
Gait Recognition
Biometrics are body measurements and calculations based on unique physical characteristics such as fingerprints, palm veins, facial features, palm prints, iris recognition, or retina. Biometrics can be used for automated recognition for identification and access control.
KeystrokeRecognition
Something you are...
Security Concerns
Performance Metrics
Quiz
Biometric Identifiers
The Mission
Fingerprint recognition is one of the oldest and most popular techniques. There are three fingerprint matching techniques: 1) Minutiae Based - points are found and mapped to the relative position on the finger; 2) Correlation Based - improves on minutiae based by being able to work with bad quality data; 3) Pattern Based - compares the fingerprint patterns (arches, whorls, and loops) to stored data.
Physiological
Voice recognition combines physiological and behavioral components. Physical shape, size, and health of a person's vocal cord plus the lips, teeth, tongue, and mouth cavity make up the phsiological component.
Hand geometry includes measuring length and width of the palm, surface area, length and position of fingers, and overall bone structure of the hand. Hand geometry is not unique, though, so it is not as reliable. Hand geometry is more effective in adults, but not for growing children.
Physiological characteristics are related to the shape of the body and measures similarity, not identity. A biometric system compares characteristics to one or more previously recorded references. If a characteristic is suitably similar, it is from the same individual. Therefore, the individual is recognized as someone previously known to the system.
Click each of the highlighted icons to learn more.
DNA is the genetic material found in the nucleus of a cell. DNA patterns can be collected from blood, saliva, nails, and hair. DNA recognition is highly accurate, but it requires a physical sample.
Facial recognition uses the shape and size of the jaw, chin, shape and location of the eyes, eyebrows, nose, lips and cheekbones. It can be performed several ways: 1) Facial Metrics use the distances between pupils or from nose to lip or chin; 2) Eigen Faces analyzes the overall face image as a weighted combination of a number of faces; 3) Skin Texture Analysis locates the unique lines, patterns, and spots in a person's skin.
Finger geometry captures features such as the shape and surface area of each finger, its length, width, thickness, and distance between fingers. It provides a usable identity process for large numbers of users where identity assurance or security requirements are low since these shape measurements do not provide a unique biometric in the same way as a fingerprint would.
Vein recognition (also known as vascular biometrics) measures a person's circulatory system which is as unique as a fingerprint. Optical scanning technology captures vein images in the palm, finger, or eyeball. Vein patterns are virtually un-spoofable.
Iris recognition uses the iris pattern, which is the pigmented elastic tissue in the eye with an adjustable circular opening in the center. In adults, the texture of the iris is stable throughout their lives. The iris patterns of the left and right eyes are different.
The retina is the layer at the back of the eyeball the makes up 65% of the inner surface. It contains photosensitive cells, and each person's retina is unique becuase of the complex network of blood vessels that supply blood. The retina pattern remains unchanged throughout a person's life except in the case of diabetes, glaucoma, or other degenerative disorders.
Biometric Identifiers
Physiological
Behavioral
Good Design Principles
Performance Metrics
Behavorial biometrics identify measurable patterns in human activities. Behavioral biometrics hare more variations since they depend on external factors such as fatigue or mood.
Click each of the highlighted icons to learn more.
Signature recognition includes changes in the timing of writing, pauses, pressure, direection of strokes, and speed during the signature. The signature recognition algorithm assigns different values of weights to these features in addition to important points on the path of the signature and the size of the signature.
The behavioral component of voice recognition is the emotional status of the person while speaking. This includes accents, tone, pitch, pace of talking and even mumbling.
Behavioral
This biometric analyzes a person's typing pattern which is the rhythm and the speed of typing on a keyboard. It is based on the dwell time, the duration of time a key is pressed, and flight time, the time elapsed between releasing a key and pressing the next key.
Gait is the manner of a person walking. Body posture, the distance between two feet while walking, and swaying are all traits that help to recognize a person. A sample of an individual's walk cycle records and analyzes the position of the knees and ankles and the angles between them while walking. The stored sample is compared with the live walk sample.
Click here to scan
The FMR (or False Accept Rate, FAR) is the probability that the system incorrectly matches the input pattern to a non-matching template in the database. It measures the percent of invalid inputs that are incorrectly accepted. In case of similarity scale, if the person is an imposter in reality, but the matching score is higher than the threshold, then he is treated as genuine. This increases the FMR, which also depends upon the threashold value.
False Match Rate (FMR)
Template Capacity
Equal Error Rate
False Match Rate
Failure to Enroll Rate
Failure to Capture Rate
False Non-match Rate
False Non-match Rate
Biometrics
The False Non-match Rate, FNMR, (or False Reject Rate, FRR) is the probability that the system fails to detect a match between the input pattern and a matching template in the database. It measures the percent of valid inputs that are incorrectly rejected.
Equal Error Rate (EER)
Least Privilege
The EER (or crossover error rate, CER) is the rate at which both false acceptance rates (FAR) and false rejection rates (FRR) are equal. The value of the EER can be easilty obtained from the ROC curve which is a graphical plot of error rates. The EER is a quick way to compare the accuracy of devices with different ROC curves. The device with the lowest EER is the most accurate.
Template Creation
The FTE (or FER) is the rate at which attempts to create a template from an input is unsuccessful. This is most commonly caused by low-quality inputs.
Feature Extraction
Fingerprint Scan
Technical Challenges
If biometrics are compromised or stolen, stolen identity credentials can be used for any kind of theft or falsification. Additionally, an individual's biometrics cannot be replaced.
Biometrics can be expensive to implement from scanners to servers that store data.
Biometric recognition mistakes include bias in training data, denial of entry due to erroneous scanning, a compromise of the biometric (a cut finger, for instance), or false positives and false negatives.
Legal Concerns
US laws around use of biometric data are inconsistant among states.
Emerging technologies always have unintended consequences when implemented in the real world.
Newer digital interfaces can be used in a passive way that does not require the participation of the individual.
Biometrics have the advantage from a security perspective. The security factor increases when combined with other methods such as PINs or security questions. Biometrics are difficult to hack and are challenging to replicate. Despite these advantages, there are several concerns that an organization needs to evaluate in its due diligence.
Click on each icon to learn more
Ethical Issues
Biometric data collection opens up the question of its use beyond access control since there are a number of entities buying and selling consumer data.
Questions surrounding citizenship, surveillance, and human rights can be exploited.
The probability that the system incorrectly matches the input pattern to a non-matching template in the database.
The rate at which attempts to create a template from an input is unsuccessful.
Quiz 1 of 5
Biometrics
Failure to Capture
The probability that the system fails to detect a match between the input pattern and a matching template in the database
Element
with Audio
HTML
Failure to Enroll
Great Job!
The probability that the system fails to detect a biometric input when presented correctly.
Drag each term to its appropriate description.
A quick way to compare the accuracy of devices with different ROC curves.
The Mission
Facial scanners determine the face geometry based on the shape and size of the jaw and chin, the shape and size of the eyes, eyebrows, nose, lips, and cheekbones. A database stores the facial geometry in terms of points. A comparison algorithm performs face matching to come up with the results.
That’s correct!
A system should be subdivided into smaller parts that can be independently created and then used in several systems.
Close
Iris Recognition
Eyebrow Recognition
That’s correct!
Quiz 2 of 5
Cybersecurity Design Principles
Finger Geometry Recognition
Facial Dynamics
Bob wants to implement a biometric solution that uses physiological characteristics. Which of the follow options will accomplish Bob’s goal?
Select all possible answers.
Retina Recognition
Quiz 3 of 5
Which of the following methods measures a biometric characteristic that is as unique as an individual’s fingerprint?
Voice Recognition
Vein Pattern Recognition
Keystroke Recognition
False acceptance rate and fale match rate are equal.
False acceptance rate and false rejection rate are equal.
The Equal Error Rate is the rate at which _____________________.
The template capacity equals the failure to capture rate.
False match rate and False nn-match rate are equal.
Quiz 4 of 5
Signature
Gait
Behavioral
Both Physiological
and Behavioral
Vein Pattern
Keystroke
Drag the biometric characteristic into its proper category.
Quiz 5 of 5
DNA
Fingerprint
Physiological
Voice
Restarting
Stuxnet
Failure to Capture Rate (FTC)
Within automatic systems, the probability that the system fails to detect a biometric input when presented correctly.
Different biometric systems have different capacity to store biometeric templates. Template capacity tells the maximum number of sets of data that can be stored in the system.
Template Capacity
Every person has a unique fingerprint composed of ridges, grooves, and direction of the lines. Ridges form three basic patterns—the arch, loop, and whorl. The uniqueness of a fiingerprint is determined by the pattern as well as ridge endings called bifurcation and spots.
Click here to enter
Iris recognition works on the basis of the iris pattern of the eye. The texture of the iris is stable throughout one’s life. The iris patterns of the left and right eyes are different. A camera takes a picture of the iris, stores it, and compares it to a stored image using a mathematical algorithm. An iris scan will work if the person is wearing glasses or contacts.
The objective of voice recognition is to identify WHO is speaking. The user enrolls by speaking the phrase into a microphone to acquire the speech sample. The speech sample is converted into a digital signal and stored. When the user provides the live input sample, the system compares it to the stored digitized sample for verification.
Welcome to the secret laboratory!
Did you bring
the microchip?
Jackson is on a mission to deliver a microchip containing the secret formula needed
to produce a vaccine that will save the world.