Functioning of a Biometric System

Biometrics is fundamentally a multi-disciplinary area of research, which encompasses subjects like pattern recognition or deciphering patterns, digital image processing, computer vision, soft computing, and artificial intelligence. For instance, face image is acquired by a digital camera, which is pre processed using image enhancement algorithms, and then facial information is extracted and matched. Have described the various pattern recognition algorithms which are applicable to various biometric techniques which are illustrated in Figure 6.

Throughout this procedure, image processing methods are used to develop the face image and pattern recognition, and soft computing techniques are used to extract and match facial features. A biometric system can serve as an identification system or a verification (authentication) system. Biometric jargon such as “verification” and “identification,” are used interchangeably in some books. This erroneous overlap creates misunderstanding as each term has a specific definition. A concise description of these important terms is provided here.

Biometric authentication is carried out by matching the biometric characteristics of an enrolled individual with the features of the query subject. Several phases of a biometric system are: capture, enhancement, feature extraction, and matching. Throughout capture, raw biometric data is acquired by an appropriate device, such as a fingerprint scanner or a camera. Enhancement encompasses developing the raw data to enhance the quality of the data for exact feature extraction. Enhancement is particularly necessary when the quality of the raw data is not good—for instance, if the face image is unclear or contains noise. The raw data includes plenty of superfluous information which is not helpful for recognition. Feature extraction encompasses extracting invariant characteristics from the raw data and generating biometric template which is distinctive for every individual and can be employed for recognition.