Image Forgery
Detection
PD
Presented by:
Name of Participant Enrollment no Division
Samruddhi Pise 2207100181 B
Diksha Dhomse 2207100671 B
Priya Thorat 2207100775 C
Tanuja Dhobale 2207100791 C
Contents
INTRODUCTION
WHY TOPIC CHOOSEN
FORENSIC IMAGE ANALYSIS
ALGORITHM
FLOWCHART
BLOCK DIAGRAM
CONCLUSION
REFFERENCES
Introduction
Image forgery detection involves identifying and analyzing alterations
in digital images to determine their authenticity. With the proliferation
of image editing tools, distinguishing between genuine and
manipulated images has become increasingly important.
Image Forgery Detection is the process of identifying whether an image
has been manipulated or altered using various tampering techniques
such as splicing, copy-move, or image retouching. With the increasing
accessibility of photo-editing software, the need for automatic and
reliable methods to detect image forgeries has become critical.
.
Why This Topic is Chosen?
Increasing prevalence of image manipulation.
Misinformation Financial Fraud Legal Disputes
Resolving disputes involving
Spreading false information. Using fake images for scams.
image evidence.
Importance of Image
Authenticity
Ensures trust and credibility in online content.
1 News and Media 2 Social Media
Preventing Combatting fake
manipulation of news and
events. misinformation.
3 Legal Evidence
Ensuring authenticity in legal proceedings.
Techniques for Image Forgery Detection
Various methods to detect manipulated images.
Pixel Analysis Geometric Analysis Statistical Analysis
Analyzing pixel values for Examining geometric Analyzing image statistics for
inconsistencies. distortions and anomalies.
inconsistencies.
Forensic Image Analysis
In-depth analysis of image data for evidence.
1 Image Acquisition
Collecting and preserving image data.
2 Image Processing
Enhancement and analysis of image features.
3 Report Generation
Documenting findings and conclusions.
Lighting and Shadow
Inconsistencies
Analyzing inconsistencies in lighting and shadows.
Light Source Direction
Should be consistent throughout the image.
Shadow Lengths
Should be proportional to the objects.
Shadow Casts
Should align with the light source.
Algorithm:
Step 1 : Input Image: Load the image.
Step 2 : Preprocessing: Convert to grayscale and resize.
Step 3 : Feature Extraction:
Pixel-level analysis (detect abrupt changes).
Error Level Analysis (ELA) (analyze compression differences).
Statistical analysis (mean, variance, etc.).
Step 4 : Forgery Detection:Copy-Move: Detect duplicated [Link]: Detect boundary inconsistencies.
Step 5 : Decision: Combine results from the detection techniques.
Step 6 : Output: Highlight tampered areas or return "authentic."
Start
Flowchart: Input Image
Preprocessing
Feature Extraction
Copy-Move Detection
Splicing Detection
Forgery Decision
Output
End
Block Diagram:
Conclusion and Future
Developments
Constant evolution of techniques to combat forgeries.
Real-Time Detection Advanced Algorithms
Detecting forgeries in Developing more
real-time applications. sophisticated
detection algorithms.
Cross-Platform Detection
Extending detection to various platforms and formats.
References
Sources used for research and information.
• Journal of Forensic Sciences
• IEEE Transactions on Information Forensics and Security
• Computer Vision and Image Understanding
Thank you!!