A DEEP FACIAL

RECOGNITION SYSTEM
USING COMPUTATIONAL INTELLIGENT
ALGORITHMS

SUBMITED BY-
SUBMITTED TO – Himanshi Sharma(22/297)
Prof. Harish Sharma Jangid Diya(22/299)
HOD, Computer Science Department, Komal Agrawal(22/304)
RTU , Kota Nishu (22/322)
 CONTENTS:
1. ABSTRACT
2. INTRODUCTION
3. FR TECHNOLOGIES AND APPLICATIONS
4. COMPONENTS OF TRADITIONAL FR
5. COMPUTING FRAMEWORK
6. DEEP CONVOLUTIONAL NEURAL NETWORKS
(DCNN)
7. COMPARITIVE ANALYSIS
8. RESULTS AND DISCUSSIONS
9. CONCLUSION
10. REFERENCES
 ABSTRACT

• The FRT is a biometric technology .

• Used t o an individual and analyse the behaviour.
• Using facial features with the help of software.
• The FR system uses DEEP CONVOLUTIONAL NEURAL NETWORKS
(DCNN).
• DCNN is used to extract relevant facial features.
• Features compare faces in an efficient way.
• The proposed recognition method (DCNN) was tested with different
three
standard machine learning algorithms.

1. Decision T ree (DT)

2. K Nearest Neighbour(KNN)
3. Support Vector Machine (SVM)
INTRODUCTION
• The FR system has vast benefits compared to other biometric
security solutions such as palmprints and fingerprints.
• There are approximately 80 to 90 unique nodal points of a face.
• From these nodal points, the FR system measures significant
aspects.
1. the distance between the eyes,
2. length of the jawline,
3. shape of the cheekbones, and
4. depth of the eyes

• These points are measured by creating a code called the faceprint,

• A digitally recorded representationof person’s face ,
• as individual as fingerprint.
 FR TECHNIQUES AND APPLICATIONS:

• Criminals identifications.
• Law enforcement use at railway stations.
• Passenger check-ins at airports .
• Biometric attendance at companies and schools.
• as a tool to identify lost people.
• T o identify people who repeatedly turned up at
protests.
• Betterborder controls and countering terrorism, etc.
THE COMPONENTS OF TRADITIONAL FR:

The whole system comprises three modules:

a. Face detection - the face detector ,utilized on videos or images to
detect faces.
b. Face modeling - feature detector aligns each face to be normalized
and recognized with the best match.
c. Face feature extraction - the face images are fed into the FR module
with the aligned results.

Before inputting an image into the FR module, the image is scanned using
face anti-spoofing .
Support vector machine

• SVMs are among the best off-the-shelf supervised learning models that are
capable of effectively working with high-dimensional datasets.

• Creates a decision boundary that segregate n-dimensional space into classes.

• can easily put the new data point in the correct category in the future.

• SVMs are useful in several real-world systems including:

- protein classification,
- image classification, and
- handwritten character recognition
COMPUTING FRAMEWORK
• the computing framework is one of the essential parts for processing data.
• The computing framework is famous for cloud and fog computing.
• The application of FR can utilize a framework based on process location
and application.

1. Fog computing -
• Cloud computing transmits information to the edge of the servers from the
datacenter task.
• The fog computing architecture on edge servers uses this architecture
• it provides network, storage space, limited computing, and data filtering of
logical intelligence and datacenters.
• This structure is used in fields like military and e-health applications.
 Cloud computing -
• Toobtain accessible data, data are sent to the datacenter for analysis and
processing.
• Big data processing increases the cloud server’s CPU usage .
• There are various types of cloud computing such as
- Infrastructure as a Service (IaaS),
- Platform as a Service (PaaS),
- Software as a Service (SaaS), and
- Mobile Backend as a Service (MBaaS).
Traditional deep convolutional neural
networks
• Images are expressed in terms of width (W) 227, height (H) 227, and depth (D)
3. of the colors red, green, and blue.
• they have a size of 227×227×3.
• The input color image is filtered at the first convolutional layer.
• This layer has 96 kernels (K) with an 11x 11x11 filter (F) and a 4-pixel stride (s).
• The second input of the convolutional layer has a size of 55×55×no of filters,
and therefore, the number of filters is 256 in this layer.
• As the work of the layers is distributed over 2 GPUs, the load is divided by 2
over all layers in each GPU.
• The next layer is the convolutional layer, followed by the pooling layer.
• The pooling method can be max, sum, average , etc. A max-pooling layer
is employed in AlexNet.
• AlexNet comprises five Conv layers (CL1, CL2, CL3, CL4, and CL5)
and three completely associated layers (FCL6, FL7, and FL8).
• The normalized output of the second convolutional layer is connected to the
third
layer, which has 384 kernels.
• The last three convolutional layers are created without pooling layers or
normalization.
The outputs of these three layers are delivered as the input to two fully
connected
layers, where each layer has 4096 neurons.
AlexNet
architecture

https://doi.org/10.1371/
Adaptive deep convolutional neural networks (the
proposed face recognition system)
The proposed system consists of three essential
stages-
1. preprocessing-
• the frame begins to capture images that must have a human face
as
• to
thedetect faces
subject in a specific image, this face detector uses detection windows
of insertion.
with
different sizes to scan the input image.

2. feature extraction-

•the preprocessed ROI in handled to extract feature vector using the modified
version
of AlexNet.
3. recognition, and
identification-
the determination of feature vector belongs to whom subject of enrolled subject
in
the system’s database.
• for the feature vector of already register subject, the system recognition
the associated ID.
• for the feature vector of a new registered subject, the system adds new record
into
the connected database.
Results & discussion:
After the comparison of four algorithms using the three datasets SDUMLA-HMT,
113, and CASIA ,The results show that the proposed DCNN has superiority
over other machine learning algorithms according to observation/sec and
recognition time , in Precision results, according to Recall parameters, Accuracy
parameters and in Specificity results.
Recognition time results-

doi:
 Average
results

doi:
doi:
 CONCLUSION
• This study is based on the adaptive version of the most recent DCNN
algorithm, called AlexNet.
• In DCNN steps are divided by preprocessing, face detection, and feature extraction.
• In this study, DCNN,DT, KNN, and SVM algorithms, were tested on three standard
benchmark datasets.
• The results showed that the DCNN could be used to effectively locate an
optimal solution within a reasonable time compared with other popular
algorithms.
• In future , there is two ways to improve the algorithm-

1. by comparing the proposed algorithm with different recentmetaheuristic algorithms

and testing the methods with the remaining instances from each dataset.
2. by applying the proposed algorithm to real-life FR problems in a specific domain.
REFERENCES

• Schiller Dominik, Huber Tobias, Dietz Michael, and Andre´ Elisabeth.

"Relevance-based data masking: a model-agnostic transfer learning
approach for facial expression recognition." (2020).
• Prakash, R. Meena, N. Thenmoezhi, and M. Gayathri. "Face Recognition
with Convolutional Neural Network and Transfer Learning." In 2019
International Conference on Smart Systems and Inventive Technology
(ICSSIT), pp. 861–864. IEEE, 2019.
• Cristianini N, Shawe-Taylor J. An introduction to support vector machines
and other kernel-based learning methods. Cambridge, UK: Cambridge
University Press; 2000.
 Important Questions:
❑ Very Short Answer Types:

1) What is face recognition?

Ans: Face recognition, also known as facial recognition, is a technology and biometric method
used to identify or verify individuals by analyzing and comparing their facial features.

2) What is DCNN?
Ans: Deep convolutional neural networks (CNN or DCNN) are the type most commonly used to
identify patterns in images and video.

3) What do you mean by supervised learning?

Ans: It is defined by its use of labeled datasets to train algorithms that to classify data or predict
outcomes accurately.
4) Definition of Cloud Computing?
Ans: Cloud computing is the delivery of different services through the Internet. These
resources include tools and applications like data storage, servers, databases, networking,
and software.

5) Define Decision Tree?

Ans: A decision tree is a flowchart-like tree structure where each internal node denotes the
feature, branches denote the rules and the leaf nodes denote the result of the algorithm .

❑ Short Answer Types:

1) What are the components of Traditional FR?

Ans: The whole system comprises three modules:
a. Face detection - the face detector ,utilized on videos or images to detect faces.
b. Face modeling - feature detector aligns each face to be normalized and recognized
with the best match.
c. Face feature extraction - the face images are fed into the FR module with the aligned
results.

2) What are the application of face recognition technique?

Ans: Following are applications of face recognition technique:
• CNN is used to classify COVID-19-infected patients as infected (+ve) or not (−ve) by the
chest CT images.
• Criminals identifications.
• Law enforcement use at railway stations.
• Passenger check-ins at airports .
• Biometric attendance at companies and schools.
• as a tool to identify lost people.
• To identify people who repeatedly turned up at protests.
• Better border controls and countering terrorism, etc.
3) Explain Support Vector Machine?
Ans: SVMs are –
• among the best off-the-shelf supervised learning models that are capable of effectively
working with high dimensional datasets.
• Creates a decision boundary that segregate n-dimensional space into classes
• can easily put the new data point in the correct category in the future.
• SVMs are useful in several real-world systems including
- protein classification,
- image classification, and
- handwritten character recognition.

❑ Long Answer Type:

1) Describe computer framework in detail.

Ans: • The computing framework is one of the essential parts for processing data.
• The computing framework is famous for cloud and fog computing.
• The application of FR can utilize a framework based on process location and application.

1. Fog computing –
• Cloud computing transmits information to the edge of the servers from the datacenter
task.
• The fog computing architecture on edge servers uses this architecture.
• it provides network, storage space, limited computing, and data filtering of logical
intelligence and datacenters.
• This structure is used in fields like military and e-health applications.

2. Cloud computing –
• To obtain accessible data, data are sent to the datacenter for analysis and processing.
• Big data processing increases the cloud server’s CPU usage .
• There are various types of cloud computing such as –
- Infrastructure as a Service (IaaS),
 - Platform as a Service (PaaS),
- Software as a Service (SaaS), and
- Mobile Backend as a Service (M BaaS).

2) What are the essential stages of Adaptive Deep Convolusional Neural Networks (the
proposed face recognition system)?
Ans: The proposed system consists of three essential stages-

1. preprocessing-
• the frame begins to capture images that must have a human face as the subject of insertion.
• to detect faces in a specific image, this face detector uses detection windows with different
sizes to scan the input image.

2. feature extraction-
• the preprocessed ROI in handled to extract feature vector using the modified version of
 AlexNet.
• The extract vector represents the significant details of the associated image.

3. recognition and identification-

• the determination of feature vector belongs to whom subject of enrolled subject in
the system’s database.
• for the feature vector of already register subject, the system recognition the associated
ID.
• for the feature vector of a new registered subject, the system adds new record into the
connected database.