Pattern Recognition and Biometrics

Pattern recognition deals with identifying a pattern and confirming it again. In general, a pattern can be a fingerprint image, a handwritten cursive word, a human face, a speech signal, a bar code, or a web page on the Internet.

The individual patterns are often grouped into various categories based on their properties. When the patterns of same properties are grouped together, the resultant group is also a pattern, which is often called a pattern class.

Pattern recognition is the science for observing, distinguishing the patterns of interest, and making correct decisions about the patterns or pattern classes. Thus, a biometric system applies pattern recognition to identify and classify the individuals, by comparing it with the stored templates.

Pattern Recognition in Biometrics

The pattern recognition technique conducts the following tasks −

  • Classification − Identifying handwritten characters, CAPTCHAs, distinguishing humans from computers.

  • Segmentation − Detecting text regions or face regions in images.

  • Syntactic Pattern Recognition − Determining how a group of math symbols or operators are related, and how they form a meaningful expression.

The following table highlights the role of pattern recognition in biometrics −

Pattern Recognition Task Input Output
Character Recognition (Signature Recognition) Optical signals or Strokes Name of the character
Speaker Recognition Voice Identity of the speaker
Fingerprint, Facial image, hand geometry image Image Identity of the user

Components of Pattern Recognition

Pattern recognition technique extracts a random pattern of human trait into a compact digital signature, which can serve as a biological identifier. The biometric systems use pattern recognition techniques to classify the users and identify them separately.

The components of pattern recognition are as follows −

Components of Pattern Recognition

Popular Algorithms in Pattern Recognition

The most popular pattern generation algorithms are −

Nearest Neighbor Algorithm

You need to take the unknown individual’s vector and compute its distance from all the patterns in the database. The smallest distance gives the best match.

Back-Propagation (Backprop) Algorithm

It is a bit complex but very useful algorithm that involves a lot of mathematical computations.