In this blog, our purpose is to take an in-depth look at how biometrics works. Among multiple types of biometrics like Facial characteristics, Fingerprints, Iris pattern etc. which are fairly complex to understand, we will only focus on Fingerprints. We will start with the definition of Biometric and its types. Then will talk about how fingerprint works, where we will talk about different stages of fingerprint verification process. In the final stage of the verification process, which is Matching we will only talk about Minutaie based matching using BOZORTH3 algorithm.
Finally, we will conclude the discussion with how we can make use to this technology in our day to day life to identify ourself to any service provider without the need of username and password.
What is Biometrics
Biometrics are biological instruments that can help to identify individuals. It is defined as, the automated methods of identifying or authenticating the identity of a living person based upon one or more intrinsic physical or behavioural traits. The term
biometrics is derived from the Greek words
bio (life) and
metrics (to measure). The major use of Biometric is for identification and access control. It is also used to identify an individual in groups that are under surveillance.
Types of Biometrics
There are different forms of biometric technology. Some of them are the following.
- Facial characteristics
- Hand geometry
- Retinal pattern
- Iris pattern
Researchers claim the shape of an ear, the way someone sits and walks, unique body odours, the veins in one’s hands, and even facial contortions are other unique identifiers. In this blog, we will only focus on Fingerprint.
Human fingertips are fully formed at about seven months of fetus development and ridge configurations which form distinctive patterns. The good thing about these patterns is when fully developed, they do not change throughout the life of an individual. Print of these patterns is known as fingerprints.
A bit of history here:
- Fingerprinting first created by a British surgeon called, Dr. Henry Faulds.
- In late nineteenth century, Sir Francis Galton discovered some of the points or characteristics from which fingerprints can be identified. Hence the term “Fingerprint identification” is founded by the Galton points for the science technology.
- Fingerprint identification began its transition to automation in late 1960.
- In 1969, the Federal Bureau of Investigation (FBI) developed a system to automate fingerprint identification.
- This is first used for manual process and then it is connected to National Bureau of Standards (NBS) known as National Institute of Standards and Technology (NIST) - an organization to study the process of automating fingerprint classification, feature extraction and matching.
- In the early 1990's, NIST began developing a system to enable the electronic exchange of fingerprint records and images by law enforcement agencies.
How does fingerprint works?
In general there are two stages: enrollment and verification and/or identification. The enrollment phase creates an association between the user and the user’s biometric characteristics. In this phase, the user's biometric is captures using fingerprint scanner in the form of an image. The capture device returns an image, usually with 256 grey-levels, which consists of dark (ridges) and bright (valleys) lines. Then certain unique and identifying features extracted from a user’s fingerprint are stored at the biometric database. Verification involves validating if a claimed user is the actual user. Identification is matching an unknown user from a list of potential candidates
There are two key challenges to address here:
- Scanning fingerprint cards and extracting minutiae (minutiae refers to specific plot points on a fingerprint) from each fingerprint.
- Feature extraction, comparing and matching list of minutiae against list of large repositories of fingerprints.
The most widespread fingerprint matching approach relies on the uniqueness of a fingerprint’s prominent singular points known as minutiae. These minutiae are represented either by bifurcation or termination of ridges - look at red dots in the figure below.
Basic fingerprint image consists of ridges, valleys, cores, deltas, pores etc as shown in Fig. The ridge endings and ridge bifurication's are used for comparing two fingerprints with each other, denoted minutiae based matching.
The figure illustrates ridge ending and ridge bifurcation which plays a vital role in fingerprint detection known as real minutiae. These ridge ending and ridge bifurcation do not change over time, therefore well suited for fingerprint matching. Fingerprint usually consists of 40 to 100 minutiae points. Different systems represents minutiae location differently.
There are various approaches of automatic fingerprint matching that have been proposed which include minutiae based approaches, and image based approaches. Minutia based approaches are the most popular ones being included in almost all contemporary fingerprint identification and verification system.
These are vairous stages of fingerprint verification process like shown in the figure below and can be broadly categoriese into the following:
- Fingerprint Capturing
- Feature Extraction
The first stage consists of fingerprint sensing which has been historically carried out by spreading the finger with ink and pressing it against a paper card and then scanned, resulting in a digital representation. This process is known as off-line acquisition and is still used in law enforcement applications. Currently, it is possible to acquire fingerprint images by pressing the finger against the flat surface of an electronic fingerprint sensor. This process is known as online acquisition.
Acquired image may contain noise that is removed in pre-processing stage and minutiae are extracted from pre-processed image.
Final stage for fingerprint matching is performed by passing minutiae patterns of the fingerprint to matcher. This matcher will produce a match score based on fingerprint matching.
Minutiae from the fingerprint image can be extracted directly from the fingerprint. However the extraction process is difficult and not accurate due to noise present in the fingerprint image. Also the intensity of the line pattern may differ between fingerprints. Due to these reasons the extracted minutiae from the fingerprint may contain both real and false minutiae.
Pre-processing operation is used for enhancing the contrast of the fingerprint image. Quality of the acquired fingerprint depends on the condition of the finger and sensor used. So pre-processing operation should be performed before extracting the minutiae from the fingerprint.
- Quality of the image can also be increased by using the filters. Low pass filter decrease the noise from the image, band pass filter decrease undesired noise from orientations which helps in preserving true ridges
- Fingerprint enhancement using Fourier Transform
- Fingerprint enhancement using Histogram Equalization (Histogram Equalization is a computer image processing technique used to improve contrast in images.)
Minutiae extraction of the fingerprint is performed by using the algorithm developed by NIST.
- Image maps are generated from the input fingerprint.
- Image binarization is performed and the minutiae are detected from the
binarized fingerprint image.
- Removal of false minutiae from the fingerprint happens.
- Finally, output minutiae file is generated after accessing minutia quality.
Minutiae extracted from the fingerprint consist of real and false minutiae. The number of falsely detected minutiae depends upon the quality of the fingerprint. These false minutiae much be filtered to remove as many false minutiae as possible without removing real minutiae. The redundant minutiae in the fingerprint are of the form:
- Minutiae Points adjacent to each other
- Minutiae near the border
- Spike, break, bridge, hole.
- a) Fingerprint image containing both false and real minutiae
- b) Fingerprint image containing real minutiae with "o" in green and false minutiae with "o" in red.
Fingerprint matching is a difficult approach due to quality variations of the fingerprint from the same user in time. These variations are due to changing skin conditions, noise, errors caused during extraction.
Types of matching are:
- genuine match : matching is from same person
- imposter matching : matching is from different persons
Some of the fingerprint matching techniques are:
- Correlation-based matching
- Minutiae-based matching
- Ridge feature-based matching
We will only talk about Minutiae-based matching since this the one very popular.
Minutiae-based matching performed using BOZORTH3 algorithm. This algorithm generates a score based on pairing minutiae of the fingerprints. BOZORTH3 is an algorithm and utility that matches two minutiae patterns with each other and produces a match score.
Bozorth algorithm includes three steps for fingerprint matching
- Construction of Intra-Fingerprint Minutiae Comparison Tables
- Construction of Inter-Fingerprint Compatibility Table
- Traverse Inter-Fingerprint Compatibility Table constructed in second step
First minutiae tables are constructed for both probe and gallery fingerprint as shown in fig below. In the table
Y represents the coordinates of each minutiae,
T represents orientation
D represents the minutiae quality. These tables containing all minutiae parameters are then passed to the minutiae matcher for generating the scores.
S(P1,P2) represents scores obtained from genuine match where
P2 are the fingerprints from same person.
S(G1, G2) represents scores obtained from imposter match where
G2 are the fingerprints from different persons. As larger the matching score, the fingerprints are more likely from the same person.
The use of authentication builds a sense of trust for the end users, prevents access to a the network or service to intruders (i.e., external attack), and is necessary for holding a person accountable, in case the threats were a result of her actions (i.e., internal attack). Authentication along with the CIA Triad (confidentiality, integrity and availability) is one of the most important aspects of any cyber-physical system.
One of the simplest ways to implement user authentication is to use single-factor authentication. The widely adopted form of single-factor is to an username and password to log on to a website which tells
Something a user knows!. It provides limited protection against attackers, hence MFA - Multifactor authentication came into picture which tells
Something the user has!. Multi-factor authentication using the first two factors definitely improves the quality of verification, but can still be vulnerable to attacks. If a malicious user steals both the password and smart card of a legitimate user, the attacker can gain access to the user’s account.
In this case, Biometrics seems to be a better option to proof user's identity since it talks about
Something the user is!