project final review VS about cryptography and network security
1. DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING 1
PROJECT FINAL REVIEW
TEAM MEMBERS:
310520104116 SHAKTHIKUMAR S
310520104302 JAYASURYA B
310520104143 YASHVANTHKUMAR V
310520104133 VASANTH
GUIDED BY:
Ms.SOWMIYA
DHANALAKSHMI SRINIVASAN
COLLEGE OF ENGINEERING AND TECHNOLOGY
2. DHANALAKSHMI SRINIVASAN
COLLEGE OF ENGINEERING AND TECHNOLOGY
DOMAIN: CRYPTOGRAPHY AND NETWORK SECURITY
TITLE: Advancements in Video Steganography with Multifactor Authentication
Using Convolutional Neural Networks
2
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
3. ABSTRACT
In the era of digital communication and data security, the need for concealing sensitive information in
plain sight has become increasingly critical. This project, titled "Video Steganography using CNN with
Password and Audio Authentication," addresses this need by developing a novel approach to hide one
video within another, secured by authentication mechanisms. The project's process is initiated when a
user submits two videos: a "cover video" and a "hide video," along with a unique password and an audio
clip for authentication. CNNs are adept at understanding spatial relationships in images and videos,
ensuring a seamless integration of the hide video. The authentication mechanism relies on two factors:
the user-supplied password and the audio clip provided. These two elements are crucial for successful
recovery of the hidden video.
3
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
4. OBJECTIVE
The objective of the "Video Steganography using CNN with Password and Audio
Authentication" project is to provide a secure and innovative method for concealing sensitive
information within digital videos.
Leveraging Convolutional Neural Networks (CNNs), the project aims to seamlessly embed a
"hide video" within a designated "cover video.“
The incorporation of CNNs ensures a robust understanding of spatial relationships in videos,
facilitating a covert integration.
The authentication process, essential for successful recovery, involves a unique user-supplied
password and an audio clip.
By combining these two elements, the project aims to enhance data security and
confidentiality in digital communication, offering a reliable solution for concealing
information in plain sight.
4
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
5. EXISTING SYSTEM
Model Technique:
Deep Convolutional Neural Network: Utilizes this technique for efficient concealment.
Suitability of Large Media Files:
Large media files are ideal due to their size for steganographic transmission.
Focus on Video Steganography:
Exploration of concealing a secret video within a cover video
DISADVANTAGES:
This process only hide the video and recovery then demonstrating backend.
There is no authentication in this existing system.
5
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
6. PROPOSED SYSTEM
Building on Video Steganography:
Enhances existing video steganography methods.
Focus on improving data concealment, security, and authenticity verification.
Use of Convolutional Neural Networks (CNNs):
Leverages CNNs for a strong and efficient framework.
Aims to hide a secret video within a cover video effectively.
Multifactor Authentication:
Password + Audio Clip: Users provide both for added security.
Essential for successfully recovering the hidden video.
Secure Storage of Passwords:
Passwords are stored securely in a database to ensure confidentiality.
Innovative Residual Modeling Technique:
Allows seamless embedding of the secret video in the cover video.
Minimizes detectable differences for enhanced concealment.
6
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
7. LITERATURE REVIEW
7
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
YEAR - TITLE METHODS ADVANTAGES DISADVANTAGES HOW THIS REFERENCE GIVES SCOPE FOR
PROPOSED WORK
2019 -Secure
Multimedia
Steganography
Using Deep
Learning
The authors
employ
a convolutional
neural network
(CNN) with
a Deep
Supervision-
based edge
detector. This
edge detector
retains more
edge pixels
compared to
conventional
edge detection
algorithms
Deep learning
algorithms can
optimize the process
of concealing
information within
multimedia files,
leading to more
effective and
seamless embedding
that is difficult to
detect.
Training effective deep
learning models for
steganography may
require large and diverse
datasets, which might not
always be available,
especially for specific
domains or types of
multimedia
This paper explores the progress in computer
vision techniques as applied to autonomous
vehicle navigation. It discusses the use of
computer vision for enhancing the perception
and decision-making capabilities of
autonomous vehicles, highlighting the
significance of advanced technologies in
ensuring safe and reliable autonomous
navigation.
8. LITERATURE REVIEW
8
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
YEAR - TITLE METHODS ADVANTAGES DISADVANTAGES HOW THIS REFERENCE GIVES SCOPE FOR
PROPOSED WORK
2020 - Audio
Authentication
Techniques for
Multimedia Data: A
Comprehensive
Survey
The authors
propose an
algorithm that
combines per
ception-
based robust
hash
functions wit
h robust
watermarkin
g.
Audio authentication
techniques can
provide robust
security measures,
ensuring the
integrity and
authenticity of
multimedia data.
Audio authentication
systems may produce
false positives
(authenticating an
unauthorized user) or
false negatives (rejecting
an authorized user),
impacting the overall
reliability of the system
This comprehensive survey explores audio
authentication techniques in the context of
multimedia data. It provides an overview of
various methods for verifying the authenticity
of audio data within multimedia content. The
paper highlights the need for robust
authentication mechanisms in multimedia
applications
9. LITERATURE REVIEW
9
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
YEAR - TITLE METHODS ADVANTAGES DISADVANTAGES HOW THIS REFERENCE GIVES SCOPE FOR
PROPOSED WORK
2019 - Data Privacy
and Security in
Multimedia
Communications:
Challenges and
Solutions
Conventional
systems inclu
de methods
like encrypti
on and water
marking,
which have
been widely
used for
securing
multimedia
data.
By examining the
progress in IoT-
related multimedia
privacy and security
research, the book
aims to provide
investigators with
a greater
understanding of
the current state-of-
the-art.
The challenges arise due
to the massive amounts of
multimedia content
exchanged via IoT
devices, particularly in
data-sensitive scenarios
This paper discusses the challenges and
solutions related to data privacy and security
in multimedia communications. It addresses
the evolving landscape of multimedia data and
emphasizes the importance of ensuring the
privacy and security of sensitive information
during transmission.
12. ALGORITHM
12
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
The "Video Steganography using CNN with Password and Audio Authentication" project, Convolutional
Neural Networks (CNNs) play a pivotal role in achieving effective video concealment and security. CNNs
are a class of deep neural networks particularly well-suited for image and video processing tasks. In this
project, the CNNs are utilized to comprehend and manipulate the spatial relationships within the cover
and hide videos. The network is trained to understand the intricate patterns and features present in video
frames, allowing for the seamless integration of the hide video into the cover video.
Specifically, the CNN acts as a powerful tool for feature extraction, capturing relevant spatial information
and patterns in both videos. Through a process of convolutional layers, the network learns hierarchical
representations of the input videos, enabling it to discern subtle details crucial for concealing the hide
video. The application of CNNs in this project ensures that the embedding process is efficient and
minimizes detectable differences between the cover and modified videos.
The use of CNNs enhances the system's capability to conceal information within the videos effectively. By
leveraging the neural network's ability to grasp spatial relationships, the project achieves a high level of
integration between the cover and hide videos, contributing to the overall success of the steganography
process
13. MODULES (IF APPLICABLE)
13
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
Video Input and Preprocessing
Authentication Data Handling
Convolutional Neural Network (CNN) Embedding
Authentication Process
Recovery of Hidden Video
Database Management
14. REFERENCES
[1] Patel, A. P., & Gupta, S. K. (2019). Secure Multimedia Steganography Using Deep Learning.
Journal of Information Security and Cybersecurity, 15(4), 489-503
[2] Kim, H., & Lee, C. (2020). Audio Authentication Techniques for Multimedia Data: A
Comprehensive Survey. International Journal of Signal Processing and Communication, 27(1),
89-105
[3] Wang, X., & Chen, Y. (2019). Data Privacy and Security in Multimedia Communications:
Challenges and Solutions. IEEE Transactions on Information Forensics and Security, 14(6), 1457-
1472.
14
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
![REFERENCES
[1] Patel, A. P., & Gupta, S. K. (2019). Secure Multimedia Steganography Using Deep Learning.
Journal of Information Security and Cybersecurity, 15(4), 489-503
[2] Kim, H., & Lee, C. (2020). Audio Authentication Techniques for Multimedia Data: A
Comprehensive Survey. International Journal of Signal Processing and Communication, 27(1),
89-105
[3] Wang, X., & Chen, Y. (2019). Data Privacy and Security in Multimedia Communications:
Challenges and Solutions. IEEE Transactions on Information Forensics and Security, 14(6), 1457-
1472.
14
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING](https://image.slidesharecdn.com/projectfinalreviewvs-250318165401-5cd072ff/85/project-final-review-VS-about-cryptography-and-network-security-14-320.jpg)
