Machine Learning Algorithms For Biometrics Fusion

The document discusses 5 journal articles on different biometric algorithms for fingerprint, face, iris, and palm print recognition. It explores how each algorithm used machine learning and databases to enroll and identify users, with programming languages like C++ used. Integrating such algorithms could help create more accurate multimodal biometric security systems.

Uploaded by

jasonflores2000.jf

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

48 views

Machine Learning Algorithms For Biometrics Fusion

Uploaded by

jasonflores2000.jf

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

You are on page 1/ 5

1.

Using the internet search engine, search for at least five (5) journals for different fingerprints
or any biometric algorithm and identify the following:
- How was this algorithm used as a biometric?
- How this algorithm enrolled and retrieve data?
- What programming language/s was used for this biometric?
- To where the algorithm was integrated or can be integrated?

2. Write a summary of findings about your findings and learnings

1.1 Four Machine Learning Algorithms for Biometrics Fusion: A Comparative Study
How was this algorithm used as a biometric?
- For creating a multimodal biometrics security system.
How this algorithm enrolled and retrieve data?
- The algorithms were Gaussian Mixture Models, an Artificial Neural Network, a Fuzzy
Expert System, and Support Vector Machines. The algorithms were trained and tested
using a well-known biometric database which contains samples of face and speech and
similarity scores of five face and three speech biometric experts. The fusion results were
compared against existing fusion techniques and also against each other, showing that the
fusion schemes presented in this paper produce better biometric accuracy than
conventional methods. Of the four algorithms, the most efficient one proved to be the
support vector machines-based one offering significant performance enhancement over
unimodal biometrics, over more traditionally combined multimodal biometrics, but also
over the SoA.
What programming language/s was used for this biometric?
- They used a database for this algorithm.
To where the algorithm was integrated or can be integrated?
- There were four types of algorithms that were used in this study. It is a comparison of
each model to know what’s the ideal biometric to use for better accuracy and
conventional methods.
1.2 Design and Implementation of a Multi-Modal Biometric System for Company Access
Control
How was this algorithm used as a biometric?
- The biometric features considered are the face and iris. This formula served as a
the installation of a multi-modal biometric system to control access to a corporate organization.
How this algorithm enrolled and retrieve data?
- Face is feasible for non-intrusive checking with a minimum cooperation from the subject,
while iris supports very accurate recognition procedure at a higher grade of invasively.
The recognition of the face trait is based on the Local Binary Patterns histograms, and the
Doughman’s method is implemented for the analysis of the iris data. The recognition
process may require either the acquisition of the user’s face only or the serial acquisition
of both the user’s face and iris, depending on the confidence level of the decision with
respect to the set of security levels and requirements, stated in a formal way in the
Service Level Agreement at a negotiation phase. The quality of the decision depends on
the setting of proper different thresholds in the decision modules for the two biometric
traits. Any time the quality of the decision is not good enough, the system activates
proper rules, which ask for new acquisitions (and decisions), possibly with different
threshold values, resulting in a system not with a fixed and predefined behavior, but one
which complies with the actual acquisition context. Rules are formalized as deduction
rules and grouped together to represent “response behaviors” according to the previous
analysis. Therefore, there are different possible working flows, since the actual response
of the recognition process depends on the output of the decision-making modules that
compose the system.
What programming language/s was used for this biometric?
- They used a database for this algorithm.
To where the algorithm was integrated or can be integrated?
- This algorithm was integrated for better accuracy of the said biometric.
1.3 Accuracy enhancement of biometric recognition using iterative weights optimization
algorithm
How was this algorithm used as a biometric?
- This biometric algorithm used face recognition.
How this algorithm enrolled and retrieve data?
- The importance of correlation between the overall detection performance and the area
under the genuine acceptance rate (GAR) versus false acceptance rate (FAR) graph,
commonly known as receiver operating characteristics (ROC) is recognized. Using the
ROC curve, relation between GARmin and minimum recognition accuracy is derived,
particularly for high security applications (HSA). Finally, effectiveness of any binary
recognition system is predicted using three important parameters, namely GARmin, the
time required for recognition and computational complexity of the computer processing
system. The palm print (PP) modality is used to validate the theoretical basis. It is
observed that by combining different useful feature-extraction techniques, it is possible to
improve the system accuracy. An optimum algorithm to appropriately choose weights has
been suggested, which iteratively enhances the system accuracy. This also improves the
effectiveness of the system.
What programming language/s was used for this biometric?
- They used a database for this algorithm.
To where the algorithm was integrated or can be integrated?
- For accuracy enhancement of biometric recognition.

1.4 A deep facial recognition system using computational intelligent algorithms

How was this algorithm used as a biometric?
- It uses a facial recognition biometric system
How this algorithm enrolled and retrieve data?
- Using transfer learning in fog computing and cloud computing, the developed system
uses deep convolutional neural networks (DCNN). Because of the dominant
representation; there are some conditions including occlusions, expressions,
illuminations, and pose, which can affect the deep FR performance. DCNN is used to
extract relevant facial features. These features allow us to compare faces between them in
an efficient way. The system can be trained to recognize a set of people and to learn via
an online method, by integrating the new people it processes and improving its
predictions on the ones it already has. The proposed recognition method was tested with
different three standard machine learning algorithms (Decision Tree (DT), K Nearest
Neighbor(KNN), Support Vector Machine (SVM)). The proposed system has been
evaluated using three datasets of face images (SDUMLA-HMT, 113, and CASIA) via
performance metrics of accuracy, precision, sensitivity, specificity, and time. The
experimental results show that the proposed method achieves superiority over other
algorithms according to all parameters. The suggested algorithm results in higher
accuracy (99.06%), higher precision (99.12%), higher recall (99.07%), and higher
specificity (99.10%) than the comparison algorithms.
What programming language/s was used for this biometric?
- It uses machine learning.
To where the algorithm was integrated or can be integrated?
- For more natural biometric information process than other proposed systems, and it must
address more variation than any other method.

1.5 Fingerprint Identification using Bozorth and Boyer-Moore Algorithm

How was this algorithm used as a biometric?
- The algorithm is used for fingerprint identification.
How this algorithm enrolled and retrieve data?
- A fingerprint is the ridges and valley pattern which has unique fingerprints. The
uniqueness of a fingerprint is exclusively determined by the local ridge characteristics
and their relationships. The minutiae are prominent local ridge characteristics. The
minutiae consist of two-part such as the ridge ending and the ridge bifurcation. The ridge
ending is defined as the point where a ridge ends abruptly. The ridge bifurcation is
defined as the point where a ridge forks or diverges into branch ridges. A good quality
fingerprint typically contains about 40-100 minutiae. Three patterns of the fingerprint
used in this simulation, which is a regular pattern, a pattern half erased, and a rotated
pattern.
What programming language/s was used for this biometric?
- The programming language used for Bozorth and Boyer-Moore Algorithm is C++.
To where the algorithm was integrated or can be integrated?
- The Bozorth and Boyer-Moore Algorithm can be integrated into researchers who study
fingerprint identification.

2. Write a summary of findings about your findings and learnings.

- It's true that our environment is growing more techy by the day, therefore we must adjust. the
environment quickly. With the technologies we make today and the algorithms it contains for higher
accuracy in studies from various academics, AI is becoming our future. No system is flawless, but with
the technology we now have, it is conceivable to create at least an adaptive system that will meet the
demands of our age. Any biometric technology has advantages and disadvantages, but it is up to the user
to decide how to best use it to create a reliable system.