The Global Facial Recognition Market is currently witnessing impressive growth thanks to the development of 3D facial recognition technology, and is expected to grow at a CAGR of 25.7 percent during the period 2014-2019.
3D Facial Recognition
3D face recognition is a technology that uses the 3D geometry of an individual’s face. The 3D technique uses 3D sensors to capture information such as the outline of the eyes, lips, nose, and chin. Because the information gathered by the 3D technique is more comprehensive, the rate of accuracy is much higher than that of 2D recognition.
Also, the technology used in 3D recognition is not affected by external factors such as cosmetics or lighting. However, facial expressions remain one of the major challenges in this technique because a change in facial expression changes the geometry of the face, which then becomes difficult to verify.
Facial recognition systems use a scanner and software that converts the data into a digital form and then compares the data with a pre-recorded biometric database. Biometric readers work along the following guidelines:
- When identifying a face, facial recognition systems determine who the person is
- When verifying a face, facial recognition systems determine if the person is the same as the pre-recorded data
- Capture Image: A physical sample of a face is captured by the system for enrolment
- Feature Detection: The entire face is scanned to extract the required information
- Feature Extraction: Post detection, data unique to the sample are extracted to create a template
- Store Template/Match Template: In the case of enrolment, the extracted data are stored in the database. Post enrolment, the template is matched against the stored data
- Verification: The system then decides if the claimed identity matches the one stored in the database. If the template is matched, the verification of the person is passed, whereas the system denies the verification in the case of template mismatch
Comparison of Facial Recognition with Other Biometric Technologies
The False Recognition Rate (FRR) and False Acceptance Rate (FAR) are expressed in terms of probability, which differs in every biometric technology. The FAR is the probability that the biometric system will accept the access of an unauthorized user. On the other hand, the FRR is the probability that the biometric system will reject the access of an authorized user.
