1. MINI PROJECT WORK-BEC586 PHASE
- 2 REVIEW PRESENTATION
Holographic AI Resistant Encryption for Image Protection
Dr. Kalpana A B
Professor, PG and R & D
coordinator
Department of ECE, BIT
Under the guidance of:
Sujan A Masal 1BI22EC158
Sumukh Umesh
1BI22EC160
Y S Ishwar
1BI22EC182
Presented by:
Bangalore Institute of Technology
K.R. Road, V V Puram, Bangalore-560004
Department of Electronics and Communication
Engineering
2. INSTITUTE VISION AND MISSION
Bangalore Institute of Technology Department of ECE, BIT
Vision
• Establish and develop the Institute as the Centre of higher learning, ever abreast with
expanding horizon of knowledge in the field of Engineering and Technology with
entrepreneurial thinking, leadership excellence for life-long success and solve societal
problems.
Mission
• Provide high quality education in the Engineering disciplines from the
undergraduate through doctoral levels with creative academic and professional
programs.
• Develop the Institute as a leader in Science, Engineering, Technology, Management and
Research and apply knowledge for the benefit of society.
• Establish mutual beneficial partnerships with Industry, Alumni, Local, State and
Central Governments by Public Service Assistance and Collaborative Research.
• Inculcate personality development through sports, cultural and extracurricular
activities and engage in social, economic and professional challenges.
3. DEPARTMENT VISION, MISSION, PEOs, PSOs
Bangalore Institute of Technology Department of ECE, BIT
Vision
• Imparting Quality Education to achieve Academic Excellence in Electronics and
Communication Engineering for Global Competent Engineers.
Mission
• Create state of art infrastucutre for quality education.
• Nurture innovative concepts and problem solving skills.
• Delivering Professional Engineers to meet the societal needs
PEOs
• Prepare graduates to be Professionals, Practicing Engineers and entrepreneurs the field of
Electronics and Communication Engineering.
• Capable of competing globally in multidisciplinary field
PSOs
• PSO1: The graduates will be able to apply the Principles of Electronics and
Communication Engineering in core areas.
• SO2:An ability to use the latest hardware and software tools in Electronics and
Communication Engineering
4. Bangalore Institute of Technology Department of ECE, BIT
HOLOGRAPHIC AI RESISTANT ENCRYPTION FOR IMAGE PROTECTION
• ABSTRACT
• INTRODUCTION
• LITERATURE REVIEW
• OBJECTIVES
• EXISTING METHODOLOGY
• PROPOSED METHODOLOGY
• DESIGN AND IMPLEMENTATION
• RESULTS
• CONCLUSION
• REFERENCES
CONTENTS
5. ABSTRACT
Bangalore Institute of Technology Department of ECE, BIT
• The proposed project addresses the significant issue of unauthorized use and exploitation of
digital images by AI and other automated systems, which poses privacy and intellectual
property risks for individuals and organizations.
• Current Scenario: Image security primarily is currently relying on conventional
encryption methods (AES, RSA, or SHA hashing) to protect sensitive images. These methods
secure the data, but they do not prevent images from being processed or accessed by
unauthorized AI systems.
• Proposed Solution: The project integrates advanced cryptographic algorithms with a
unique holographic overlay generated through SHA-256 hashing and FFT transformations,
creating an encrypted image with a visually obfuscating hologram pattern.
• This approach renders images unusable by unauthorized AI systems also prevents visual
identification, ensuring privacy. The hologram is secure, covering the entire image with a
pattern that is readable only with a decryption key, making it impossible for AI to reconstruct
the original image without authorization.
6. Bangalore Institute of Technology Department of ECE, BIT
What is Image Encryption?
• Image Encryption: Secures digital images by transforming them into unreadable formats.
• Goals of Image Encryption: Confidentiality, Integrity, Authentication.
• Importance: Rising use in healthcare, defense, and social media; protection against AI
misuse. Patient images and personal images exchanged over communication networks have
been intercepted and misused by creating deepfakes.
• Image Processing: Analyzing and enhancing images for applications like feature extraction
and medical imaging.
INTRODUCTION
7. Bangalore Institute of Technology Department of ECE, BIT
How Image Encryption and Processing Work?
• Image Encryption Techniques
1. Symmetric Encryption (e.g., AES):
Uses a single key for both encryption and decryption.
Fast but requires secure key sharing.
2. Asymmetric Encryption (e.g., RSA):
Uses a public key for encryption and a private key for decryption.
More secure but slower.
3. Visual Cryptography:
Splits the image into multiple shares, revealing the original only when combined.
No complex algorithms but increases file size.
4. Pixel-Based Transformations:
Alters pixel values through scrambling or mathematical operations.
Effective but may cause visual distortion.
INTRODUCTION
8. Bangalore Institute of Technology Department of ECE, BIT
How Image Encryption and Processing Work?
• Image Processing Techniques
1.Enhancement:
Improves image quality with filters and noise reduction.
Used in medical imaging and photo enhancement.
2. Restoration:
Recovers degraded image quality due to noise or distortion.
Includes techniques like de-blurring and inpainting.
3. Compression:
Reduces file size without significant quality loss (lossy) or no loss (lossless).
Used in web and digital media.
4. Analysis:
Extracts useful information, such as edges, features, and segmentation.
Key in AI, facial recognition, and autonomous vehicles.
INTRODUCTION
9. Sl.no Paper Title Author Name
Year of
Publicati
on
Publication Abstract
1
A Novel Optical
Encryption Strategy using
Virtual Digital
Holography and
Fractional Fourier Based
Key Hiding
Romita Tiwari & Mukesh
Yadav
2023
IEEE International Students'
Conference on Electrical,
Electronics and Computer
Science (SCEECS)
Presents a novel image encryption
strategy using virtual optics, digital
holography, phase shifting, and
fractional Fourier transform (FRT)
based key hiding. The sensitive
information needed to be
transferred was converted as a
binary image and encrypted.
2
A Novel Steganographic
Approach for Image
Encryption Using
Watermarking
Preethi Nanjundan,
Devanshi Singh, Aabhas Vij
& Jossy P. George
2023
9th International Conference on
Advanced Computing and
Communication Systems
(ICACCS)
Watermarking is done in the area
of steganography. Watermarking
can be done via multiple
algorithmic techniques like
Singular Value Decomposition
(SVD) and Discrete Fourier
Transform (DFT). In this study, a
combination of this along with
AES encrypted watermarked
images has been implemented.
LITERATURE REVIEW
Bangalore Institute of Technology Department of ECE, BIT
10. Sl.no Paper Title Author Name
Year of
Publicati
on
Publication Abstract
3
A Framework for Robust
Deep Learning Models
Against Adversarial
Attacks Based on a
Protection Layer
Approach
Mohammed Nasser Al-
Andoli, Shing Chiang Tan,
Kok Swee Sim, Pey Yun
Goh & Chee Peng Lim
2024 IEEE Access
An effective framework for
enhancing Deep Learning models
against adversarial attacks. The
framework uses convolutional
neural networks (CNNs) for feature
learning, Deep Neural Networks
(DNNs) with softmax for
classification, and a defense
mechanism to identify and exclude
Adversarial Samples.
4
AI Resistant (AIR)
Cryptography
Gideon Samid 2023
Congress in Computer Science,
Computer Engineering, &
Applied Computing (CSCE)
AI and quantum-AI (QAI)
represent a totally new and
effective vector of cryptanalytic
attack. Modern AI successfully
completes browser search phrases,
so it is increasingly capable of
guessing a list of plausible
plaintexts. Constructing ciphertexts
that would provide false answers to
the AI attacker and lead them
astray.
Bangalore Institute of Technology Department of ECE, BIT
LITERATURE REVIEW
11. Sl.no Paper Title Author Name
Year of
Publicati
on
Publication Conclusion
5
Holographic
Encryption
Applications Using
Multichannel Bessel
Beam
Zhi Li, Zewei Wu, Quanli
Li, Shuai Huang, Minxing
Wang, Youlei Pu, Jianxun
Wang & Yong Luo
2023
24th International Vacuum Electronics
Conference (IVEC)
In this paper, we report a
holographic encryption system
featuring a multichannel
Bessel beam as encryption key
and phase distribution of
special characters caring
information as the output.The
feasibility of this encryption
method as well as high
security is verified by
simulation.
Bangalore Institute of Technology Department of ECE, BIT
LITERATURE REVIEW
12. The primary objectives are:
• To develop a holographic encryption system to uniquely protect each image, preventing
unauthorized use.
• To implement complex hashing and AES encryption to generate unique, computationally
intensive holographic patterns for image security.
• To enable only authorized users to view original images by implementing a secure decryption
mechanism that removes the holographic overlay.
Bangalore Institute of Technology Department of ECE, BIT
OBJECTIVES
13. Bangalore Institute of Technology Department of ECE, BIT
EXISTING METHODOLOGY
1. Cryptographic Techniques
AES, RSA: Secure but computationally intensive.
Hashing (SHA-256): Ensures integrity but doesn’t block AI access.
2. Visual Cryptography
Splits images into "shares" for reconstruction. However quality is lost, vulnerable if all shares
are obtained.
3. Watermarking & Steganography
Embeds ownership info but removable by AI.
4. Adversarial Perturbations
Alters images to mislead AI models; reversible by robust AI.
5. Deep Learning Encryption
Neural networks encrypt images; computationally expensive.
6. Holographic Encryption
Uses cryptographic overlays; secure but requires advanced computation.
7. AI-Resistant Cryptography
Designed to counter AI attacks; scalable but resource-heavy.
14. Bangalore Institute of Technology Department of ECE, BIT
PROPOSED METHODOLOGY
• Pattern Generation: A unique pattern will be generated for each image based on a hash of
the image data and a secure key. The pattern will serve as a holographic overlay.
• Encryption and Masking: Using AES encryption, the holographic pattern will be
encrypted and then combined with the image using Fourier transforms, making it nearly
impossible to remove without the correct key.
• Decryption Mechanism: For authorized users, the system will decrypt the holographic
pattern, revealing the original image. This decryption will only be possible using a key
provided securely to the user.
• Testing and Evaluation: The system will be tested for robustness against common AI-
driven image processing techniques to validate its resistance to decryption without
authorization.
16. Bangalore Institute of Technology Department of ECE, BIT
DESIGN AND IMPLEMENTATION
System Workflow
1. Pattern Generation
• Generate a unique hash (SHA-256).
• Create a holographic pattern seeded by the hash.
• Apply Fourier Transform for frequency domain representation.
2. Encryption & Masking
• Encrypt the pattern using AES.
• Overlay the pattern onto the image using Fourier Transform.
• Secure storage of the encryption key.
3. Decryption Mechanism
• Decrypt pattern with the AES key.
• Reverse Fourier Transform to reveal the image.
• Key authentication ensures secure access.
17. Bangalore Institute of Technology Department of ECE, BIT
DESIGN AND IMPLEMENTATION
System Architecture
System Architecture
• Frontend: Web interface (HTML,Python Flask, CSS, JavaScript) for
encryption and decryption
• Backend: Python-based server (OpenCV, PyCryptodome) for cryptography
and processing.
• Key Management: Secure server/cloud for encryption key storage.
18. Bangalore Institute of Technology Department of ECE, BIT
DESIGN AND IMPLEMENTATION
Testing & Evaluation
• Test against distortions (resizing, cropping, filtering).
• Validate resistance to AI-based reconstruction.
• Measure performance: encryption time, image quality, security.
• Conduct penetration tests for real-world robustness.
• TensorFlow
Validates that encrypted images can’t be processed by AI models.
• Foolbox
Simulates adversarial attacks on encrypted images to ensure robustness.
21. The proposed holographic encryption project aims to create a high-security system for digital images,
using complex hashing and encryption to generate unique, unremovable holograms. This solution
provides effective deterrence against unauthorized AI use, ensuring that only authorized users can
access the original image. By securing digital images, this project advances digital rights management
for content creators, publishers, and businesses in a digital world increasingly dominated by AI.
Bangalore Institute of Technology Department of ECE, BIT
CONCLUSION
22. Bangalore Institute of Technology Department of ECE, BIT
REFERENCES
• R. Tiwari and M. Yadav A Novel Optical Encryption Strategy using Virtual Digital Holography and Fractional
Fourier Based Key Hiding 2023 IEEE International Students' Conference on Electrical, Electronics and Computer
Science (SCEECS), Bhopal, India, 2023, pp. 1-4, doi: 10.1109/SCEECS57921.2023.10063101.
• P. Nanjundan, D. Singh, A. Vij and J. P. George, A Novel Steganographic Approach for Image Encryption Using
Watermarking 2023 9th International Conference on Advanced Computing and Communication Systems (ICACCS),
Coimbatore, India, 2023, pp. 1465-1470, doi: 10.1109/ICACCS57279.2023.10112772.
• M. N. Al-Andoli, S. C. Tan, K. S. Sim, P. Y. Goh and C. P. Lim, A Framework for Robust Deep Learning Models
Against Adversarial Attacks Based on a Protection Layer Approach, in IEEE Access, vol. 12, pp. 17522-17540,
2024, doi: 10.1109/ACCESS.2024.3354699.
• G. Samid, AI Resistant (AIR) Cryptography 2023 Congress in Computer Science, Computer Engineering, & Applied
Computing (CSCE), Las Vegas, NV, USA, 2023, pp. 145-149, doi: 10.1109/CSCE60160.2023.00028.
• Z. Li et al, Holographic Encryption Applications Using Multichannel Bessel Beam 2023 24th International Vacuum
Electronics Conference (IVEC), Chengdu, China, 2023, pp. 1-2, doi: 10.1109/IVEC56627.2023.10157543
