It is a percentage of how many times the system allows access to an unauthorized person. It is a type II error in the biometric system which can result in severe harm to the premise. It possesses great importance in cybersecurity. There should be a balance between FAR and FRR. While FRR is termed a false rejection rate which is a percentage of how many times a system rejects a legitimate user.
In addition, the false acceptance rate plays a crucial role in any biometric system. The lower the rate is, the system is performing well. However, the high rate of false acceptances is an alarming situation.
Understanding Biometric Systems and the Role of FAR
Biometric systems are primarily used in secure environments, such as buildings and offices where security is a priority. Earlier this technology was restricted only to the questions related to enrollment, but now it has gone a long way and expanded to various other types such as:
- Facial Recognition
Facial recognition is a method in which the individual is verified based on their facial features. After capturing the image, the system uses liveness detection to ensure the image is from a real, live person rather than a static image or video. It then compares the image with large datasets to find a match. If it finds a person with the same identity, it allows access.
- Voice Identification
It can be done by using the voice of an individual. The system asks the user to utter some phrases so that it can identify by comparing the user’s voice with the stored data.
- Fingerprint Verification
The fingerprints are verified by analyzing the prints and converting them into a pattern and that particular pattern is then compared to the prior stored data by the system.
- Iris Scan Verification
The iris is recognized by evaluating the patterns in it. These patterns are unique in everyone’s eye. The system compares the patterns with all the other identities and verifies or denies the user accordingly.
While looking into these identification systems, the false acceptance rate (FAR) and false rejection rate (FRR) can not be neglected. FAR is the rate of accepting unauthorized users by the system while FRR is a percentage of how many times the system rejects a legitimate user. If the system has low FAR and low FRR, then the system is considered as authentic.
How to Reduce FAR?
The biometric systems are incorporated with certain algorithms that help the system evaluate the system according to its learnings. A high false accept rate can be reduced by keeping the system up to date.
- Machine Learning
This is a technique through which the system learns to identify the individuals by comparing them with the existing datasets. In case of high FAR, the system should be incorporated with the advanced algorithms. Also, this system can form better when it has a large number of data from various individuals.
- Matching Patterns or Templates
The collected data of the individuals should be rechecked and it should be compared with gathered templates of fingerprints. It can add a layer of verification to the process. By doing this the system can make accurate decisions.
- Combined Verification Systems
The system should be combined with other biometric systems like facial and PINs or more. This will help the system think twice before allowing the access to the user. If it mistakenly verifies the illegitimate user, then the other verification method can detect the spoof before allowing access.
- Awareness
The awareness process is made for the audience before implementing a new method. The users should be provided with knowledge about what is FAR and FRR. In this way, the user will be able to use the system properly and no rejection will be counted.
- Maintenance
After regular intervals, the system should be updated. It will provide more authentic and clear results. The maintenance will also make the users realize that this system is important and is implemented for their safety.
Conclusion
If the system is experiencing high FAR, then it should be fixed as soon as possible. This can make the system slow and weak. If the system, continues to allow access to illegitimate users then there will be a greater risk of scams and fraud. The lower rate of both FAR and FRR can make the whole biometric system fair and easy to use. Otherwise, continuous verifications after one rejection may frustrate the user and in the case of FAR it can make the system vulnerable.
