Module-5
1. Discuss various data hiding techniques in text.
Answer: Data hiding in text is achieved by subtly altering a text document’s formatting, content,
or structure to embed secret information while maintaining its natural appearance.
The major techniques include:
1. Line-Shift Coding
Slightly shifts the vertical position of text lines on a page.
Encoding: A small upward shift represents a ‘0’ and a downward shift a ‘1’.
Usage: Printed text steganography.
2. Word-Shift Coding
Adjusts the spacing between words.
Encoding: Increasing or decreasing space between words represents bits.
Example: Two spaces = '1', one space = '0'.
3. Feature Coding
Alters text characteristics such as font size, boldness, italics, or capitalization to embed
data.
Example: Bold = '1', Normal = '0'.
4. Open Space Methods
Inserts extra spaces after words, sentences, or lines.
Encoding: Number or position of extra spaces conveys binary data.
5. Syntactic and Semantic Methods
Syntactic: Changes punctuation or sentence structure without altering meaning.
Semantic: Replaces words with synonyms.
Example: ‘Quick’ changed to ‘Fast’ can encode data if predetermined.
6. Mimic Functions (Innocuous Text Generation)
Generates natural-looking dummy text using statistical or grammatical models,
embedding secret data.
Usage: Covertly sends messages that appear as harmless content.
7. Null Ciphers
Hide messages within a readable text using a pattern like the first letter of each word or
sentence.
2. Explain watermarking and its intuitive and digital methods.
Watermarking:
Watermarking is a technique of embedding information into a host object such that the presence
of the information is either perceptible or imperceptible, depending on the method used. It is
primarily employed for purposes such as copyright protection, authentication, tamper detection,
and ownership verification.
Types:
1. Intuitive Watermarking
This is the conventional, physical form of watermarking found on official documents, currency
notes, certificates, and legal papers.
Characteristics:
Clearly visible mark or logo embedded into the medium.
Used for immediate visual identification of authenticity.
Examples:
The watermark on currency notes.
A company logo on official letterheads or certificates.
Limitations:
Visible watermarking can be tampered with or replicated.
Provides only superficial protection.
2. Digital Watermarking
Involves embedding digital information within multimedia content in a way that it remains
hidden during normal viewing or listening, yet can be detected or extracted when needed.
Types based on embedding domain:
a) Spatial Domain Techniques
The watermark is directly embedded by modifying pixel values or text formatting.
Simple to implement but easily affected by image processing operations.
Example: Changing the brightness or position of certain pixels to represent watermark data.
b) Transform Domain Techniques
The watermark is embedded in transformed coefficients of the cover media using
transforms like DCT (Discrete Cosine Transform), DWT (Discrete Wavelet Transform),
or FFT (Fast Fourier Transform).
More robust against compression and attacks.
Example: Embedding owner information in high-frequency DCT coefficients of an image, so
the watermark remains intact even after JPEG compression.
Applications of Watermarking:
Copyright protection in images, audio, and video.
Authentication of digital content.
Tamper detection in legal and medical records.
Persistent ownership marking in shared digital files.
3. Basic features and applications of data hiding.
Features:
Conceals the existence of information.
Maintains integrity of the cover object.
Can be fragile (sensitive to modifications) or robust (resistant to transformations).
Embeds data without noticeable change.
Applications:
Covert communication
Copyright protection
Tamper detection in sensitive data (medical records)
Digital rights management
4. Mimic functions and their applications.
Definition:
Mimic functions generate text that statistically resembles natural text but carries secret messages hidden
within.
Applications:
Covertly transmitting messages
Generating fake but natural-looking text for steganography
Concealing command messages in espionage
5. Define Innocuous Text with an Example and its Role in Data Hiding.
Definition:
Innocuous text refers to natural, ordinary-looking text that appears harmless and normal to a reader
but is actually used to carry secret information. It is a vital technique in text-based steganography
where the secret message is hidden within sentences that seem completely unrelated or casual.
This type of text ensures that the hidden message remains undetected because the carrier text
does not raise suspicion and appears like a regular message or passage.
Characteristics:
Looks like natural, meaningful text.
Designed to avoid suspicion.
Used for covertly embedding secret information.
Example:
Consider the sentence:
"Simple And Neat Designs Are Beautiful."
Now, if you extract the first letter of each word, it reads:
SANDAB
This string could represent a hidden message known only to the sender and receiver, while the
sentence itself appears innocent and harmless.
Role in Data Hiding:
1. Covert Communication:
Allows secret exchange of information via common text channels like emails or letters.
2. Used in Mimic Functions:
Generates natural-sounding text carrying hidden data.
3. Used in Null Ciphers:
Specific positions in innocent text reveal hidden messages.
4. Semantic and Syntactic Techniques:
Alters words and sentence structures without affecting readability, encoding secret data subtly.
6. Explain the LSB (Least Significant Bit) Encoding Technique with a
Suitable Example.
The Least Significant Bit (LSB) encoding is one of the simplest and most widely used
techniques in image-based steganography. It involves replacing the least significant bit of image
pixels with bits of secret data without causing a perceptible change in the image.
Working Mechanism:
1. Choose a Cover Image:
An uncompressed image format like BMP or PNG is preferred because compression algorithms
may alter pixel data.
2 .Convert Secret Data into Bits:
The secret message is converted into a stream of binary bits.
3. Embed Data Bits into LSBs:
Each bit of the secret data is embedded into the LSB of the pixel’s color components (Red,
Green, or Blue).
4. Save the Stego Image:
The modified image (stego image) is saved. Visually, it appears identical to the original.
5. Data Extraction:
To retrieve the hidden data, the LSBs of the pixels are read in the same order they were
embedded.
Example:
Original Pixel Value (8-bit binary)
Red = 10010110 (150)
Green = 01101100 (108)
Blue = 11011011 (219)
Secret Data Bits to Embed:
101
After LSB Encoding:
Red = 10010111 (151) ← LSB changed to 1
Green = 01101100 (108) ← LSB remains 0
Blue = 11011011 (219) ← LSB remains 1 (no change)
7. Discuss the Working of BPCS (Bit Plane Complexity Segmentation)
Steganography.
Bit Plane Complexity Segmentation (BPCS) is a steganography technique for embedding large amounts
of data in images without noticeable degradation by replacing complex regions of an image’s bit-planes
with secret data.
Working Mechanism:
1. Bit-plane Decomposition:
Divide the image into bit-planes, from the Most Significant Bit (MSB) to the Least
Significant Bit (LSB).
2. Complexity Analysis:
Calculate the complexity of each region in the bit-planes.
Complexity is the degree of randomness in a region — highly complex areas can mask
hidden data effectively.
3. Threshold Selection:
A complexity threshold is set (typically around 0.3 to 0.5).
Blocks with complexity above this threshold are considered suitable for embedding.
4. Data Embedding:
Replace selected complex blocks in the bit-planes with blocks of secret data.
Simple regions remain untouched to preserve visual quality.
5. Reconstruction:
Combine modified and untouched bit-planes to form the stego image.
8. Compare Lossless Data Hiding and LSB Encoding.
Feature Lossless Data Hiding LSB Encoding
100% reversible, both original and Secret data can be extracted but original
Data Recovery
secret data can be restored data cannot be fully restored
Modifies data structures or pixel Replaces least significant bits of pixel
Technique
relationships without loss values
High; suitable for sensitive Low; data lost if image undergoes
Robustness
applications (e.g., medical imaging) compression or processing
Data Hiding
Moderate High
Capacity
Image Quality Preserved entirely Slight distortion, though imperceptible
Feature Lossless Data Hiding LSB Encoding
Use Cases Legal, military, and medical images General steganography applications
Complexity More complex Simple and easy to implement
9. Discuss the Advantages and Limitations of Data Hiding in Images.
Advantages:
1. High Data Capacity:
Images can carry a large amount of hidden data without visible changes.
2. Imperceptibility:
Proper techniques like LSB make changes invisible to the human eye.
3. Simple and Flexible:
Easy to implement methods like LSB, applicable to various image formats.
4. Supports Covert Communication:
Allows secret messaging through normal-looking images.
5. Can Combine with Encryption:
Hidden data can also be encrypted for extra security.
6. Used for Watermarking and Ownership Protection:
Helps in copyright verification and tamper detection.
Limitations:
1. Lossy Compression Risk:
Formats like JPEG may distort or erase hidden data.
2. Sensitive to Image Processing:
Cropping, resizing, or filtering can damage embedded information.
3. Vulnerable to Steganalysis:
Advanced analysis tools can detect statistical changes.
4. Limited Robustness:
Simple techniques like LSB are easy to implement but fragile.
5. Complex Techniques Need More Resources:
Advanced methods like BPCS require extra processing time and power.
10. Explain the Process of Data Hiding in Text Using Innocuous Text
Techniques
Innocuous text refers to ordinary, harmless-looking text used to conceal secret information. In
steganography, it allows hidden messages to be exchanged without raising suspicion, as the text
appears natural and meaningful.
Techniques of Data Hiding Using Innocuous Text:
1. Null Ciphers:
Hiding messages by using specific letters or positions in a sentence.
Example:
"Send A New Document Soon"
→ First letters spell SANDS (secret word).
2. Mimic Functions:
Generates natural-looking text that mimics regular writing while embedding secret data.
Uses statistical models to produce text patterns similar to normal text.
3. Semantic Modifications:
Altering synonyms or sentence structures to represent hidden bits without affecting
meaning.
Example:
"He is fast." → "He is quick."
4. Syntactic Variations:
Adjusting punctuation, spacing, or word order subtly to hide data.
Example:
An extra space or comma may represent a data bit.
5. Use of Special Characters or Capitalization:
Specific capital letters, punctuation, or invisible characters can denote secret data.