I M A G E E N C RY P T I O N

U S I N G M U LT I - L AY E R E D
C RY P T O G RA P H I C T E C H N I Q U E S
 AGENDA
• Abstract
• Existing Systems
• Proposed System
• Flow Chart of Proposed Method
• Results
• Technology Used
• Test Analysis
• References
 ABSTRACT
• Encryption and decryption play a crucial role in protecting digital images
by transforming them into an unreadable format, preventing unauthorized
access while allowing only authorized users to restore them securely.

• Advanced encryption techniques, such as AES , XOR and DNA Encoding,

enhance security by providing a multi-layered approach that ensures
efficiency and resistance against modern cyber threats.

• Secure image encryption and decryption are vital in fields like military,
healthcare, and digital forensics, where protecting sensitive data from
breaches and unauthorized access is a top priority.
 EXISTING SYSTEMS
1 . Static or Hardcoded Key Usage:-
Many existing systems use fixed or manually set keys, making them
predictable
and more vulnerable to brute-force or known-plaintext attacks due to
reduced
randomness.

2 . Reversible Pixel Scrambling Techniques:-

Many systems use standard pixel scrambling methods like Arnold
Transform to
chaos-based scrambling, but these follow a predictable structure that can
sometimes
be reversed if the scrambling pattern is known.

3 . Weak Obfuscation of Binary Structure:-

 PROPOSED SYSTEMS
1 . Dynamic Key Extraction for Secure Encryption and Decryption:-
The proposed system dynamically extracts encryption keys from a
randomly
selected region of the input image, ensuring unpredictability and
eliminating the need
for static key storage.

2 . Multi-Stage Protection using AES, XOR, and DNA Cryptography:-

A layered security approach using AES ,XOR and DNA encryption
ensures
that even if one encryption layer is compromised, the remaining layers
still provide
protection, making decryption without authorization extremely difficult.

3 . Bit-Level Obfuscation and Data Verification with Key-Matching:-

 KEY GENERATION PROCESS
The key generation process extracts a random 10×10 pixel block from the input image
to serve as a cryptographic key for encryption and decryption.

Steps:
• Load the Image – The image is processed and retrieves its RGB pixel data.

• Get Image Dimensions – The width and height are determined to ensure the key
block fits within the image.

• Generate Random Coordinates – A random (X, Y) position is selected within valid

limits (0 to width-10, 0 to height-10).

• Extract the 10×10 Block – A section of 100 pixels is extracted from the chosen
coordinates.

• Store and Use the Key – The key is securely saved as an image, matrix, or binary
data. It must be protected since it is essential for decryption.
 Layer 1: Pixel Scrambling
• The original image is read and converted into a NumPy array (3D for RGB).
• A key (seed here is the block 10×10) is used to generate a deterministic
random seed.

• A 1D index array is created corresponding to the flattened pixel array.

• The indices are shuffled randomly using the seed(pseudo-permutation)

• The pixel array is flattened and reordered according to the shuffled indices.

• The scrambled image is reshaped back into the original image shape.

• The shuffled indices are saved to reverse the process during decryption.
 Layer 2: AES Encryption
• The scrambled image array is converted to raw byte data.

• A secure 256-bit AES key is generated using randomly divided by 16, the image byte

data is padded to fit AES block size requirements .

• An AES cipher object is created in GCM mode (ensures integrity and confidentiality).

• The cipher encrypts the padded data and generates a tag for verification.

• The encrypted output is a combination of the cipher’s nonce, tag, and ciphertext

(nonce + tag + ciphertext).

• This byte stream becomes the AES-encrypted image data passed to the next layer.
 Layer 3: XOR
Encryption
• The AES-encrypted byte stream is converted into a NumPy array of type
uint8.

• A key (read from a file or generated) is repeated to match the length of the
data.

• A bitwise XOR operation is applied between the AES-encrypted data and the
repeated XOR key.

• This operation adds another transformation layer, altering each byte.

 Layer 4: DNA-Based
Encryption
• The XOR-encrypted NumPy byte array is converted back into raw bytes.

• Each byte is likely converted into a binary string (8 bits).

• Binary sequences are then mapped to DNA sequences using substitution rules.
("00" = A, "01" = T, "10"=C,"1"=G).

• The entire encrypted content is transformed into a sequence of synthetic DNA-

like data.

• The Dna sequence is complemented as A<->T,C<->G.

 FLOW CHART FOR PROPOSED
Extract Key SYSTEM
from image Reverse –with stored indices
(Random10x10)

Pre processing
Descramble Pixels

Scramble Image Original Image Decrypted Image

Pixels

AES Decryption
AES Encryption Encrypted Image

XOR Encryption DNA Encoding DNA Decoding XOR Decryption

 RESULTS
(ENCRYPTION PROCESS)

Original Ke Highlighted Encrypted

Image y Image Image
 RESULTS
(DECRYPTION PROCESS)

Encrypted Image Ke Decrypted

y Image
 TECHNOLOGY USED
PA C K A G E S U S E D :
• NUMPY
• PILLOW (PIL)
• RA N D O M
• P YC RY P T O D O M E ( A E S )

S O F T WA R E T O O L S U S E D :
• VS code
 T E S T A N A LY S I S

• Key Sensitivity Analysis

• Information Entropy Analysis

• Histogram Analysis

• Correlation Coefficient Analysis

• NPCR and UACI Analysis

• Key Sensitivity
Analysis :
Key sensitivity analysis is essential in image encryption because
it ensures that only the correct secret key can successfully decrypt the
image. If the key is even slightly incorrect, the decryption process does
not reveal any meaningful information, maintaining the confidentiality
of the encrypted image.
 • Information Entropy
Analysis :
Information entropy is the measure of uncertainty associated with a
random variable
NOTE: If an encrypted image has entropy close to 8, the encryption
is strong.
 • Histogram
Analysis :
Histogram analysis is an important statistical feature of the images,
which is generally used to evaluate the performance of image encryption
systems. An image histogram shows how pixels in an image are
distributed by plotting the number of pixels at each color intensity level.

The histogram is a graphical portrayal of the distribution of pixel values

in an image.

A scrambled image ought to have a uniform distribution and diverse

histogram against the plain image.
HISTOGRAM ANALYSIS
IMAGE
 • Correlation Coefficient
Analysis :
The correlation coefficient analysis is a statistical analysis performed on an
image to determine the degree of similarity between two pixels. In general,
every pixel of the original image has a strong correlation, which implies the
value is a nearby one, and each pixel of the encoded image has a weak
correlation.
 • NPCR and UACI
Analysis :
• Number of Pixels change rate measures the % of pixels the change in
the encypted image when a single pixel in the original image is
modified

• Unified Average Changing intensity (UACI) is the rate which estimates

the average intensity of the differences between the plain image and
encoded image.
 REFERENCES
1. Kumar, R. & Bhatia, P.K. (2020). A secure image encryption technique using DNA and RSA
algorithms. Multimedia Tools and Applications, 79(23), 16375–16391.

2. Wang, X. et al. (2016). A novel image encryption scheme based on DNA encoding and
discrete wavelet transform. Signal Processing, 123, 145–161.

3. Gnanajeyaraman, G. & Karthikeyan, G. (2020). Bit-level permutation and substitution-based

image encryption scheme. Journal of Information Security and Applications, 54.

4. Arun, C. & Manjunath, T.M. (2020). A secure and efficient color image encryption using pixel
value transformation and permutation. Procedia Computer Science, 167, 2588–2596.

5. Suryavanshi, R. & Dhanajayan, M. (2023). An efficient color image encryption using modified
AES with key image-based permutation. Journal of King Saud University – Computer and
THANK YOU