Image Authentication via Feature Point Clustering and Distance Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The ease of modifying and copying digital images poses challenges for media content providers in authenticating and detecting alterations, especially in secure applications like electronic commerce and legal evidence, where verifying unaltered images across various scales and formats is difficult.
Innovation Solution
A method and system for image authentication that involves feature point extraction, clustering, and distance measurement between anchor and query images, using techniques like Fuzzy C-Means Clustering and Hamming distance, to identify possible tampered areas and determine image similarity, incorporating feature point permutation and hashing for robust authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional image comparison methods are used to verify unaltered images, then image authenticity can be maintained, but the system becomes computationally complex and inefficient when handling scaled or reformatted images
Solution Approach 1:
The patent segments the image comparison process into distinct phases: feature point extraction, clustering, and distance measurement. This segmentation allows each component to be optimized independently, reducing overall computational complexity while maintaining reliability in authenticity verification
Solution Approach 2:
The patent transforms the image authentication problem from pixel-level comparison to feature point-level comparison. By changing the parameter of comparison from raw image data to extracted feature points, the system achieves both computational efficiency and robustness against scaling and reformatting operations
2Measurement precision
If feature point extraction and clustering are performed to identify tampered areas, then tampering detection accuracy is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent extracts only the essential feature points from images rather than processing the entire image data. This extraction approach maintains high tampering detection accuracy by focusing on discriminative features while significantly reducing processing time and computational resource requirements
Solution Approach 2:
The patent applies partial action by performing clustering only on extracted feature points rather than all image pixels. This selective processing achieves sufficient tampering detection accuracy without the excessive computational burden of processing complete image data
3Adaptability or versatility
If image authentication is performed on scaled or reformatted images, then adaptability to different formats is improved, but the difficulty of detecting alterations increases
Solution Approach 1:
The patent employs feature point extraction methods that are universal across different image formats, scales, and resolutions. The extracted feature points maintain their discriminative properties regardless of formatting changes, enabling the system to adapt to various formats while maintaining alteration detection capability
Solution Approach 2:
The patent introduces feature points as an intermediary representation between the original image and the authentication process. These feature points serve as format-invariant mediators that preserve essential image characteristics while being robust to scaling and reformatting, thereby facilitating alteration detection across different formats
Data Source
AI summary
In accordance with an embodiment, a method of authenticating images includes electronically receiving an anchor image and a query image, performing a feature point extraction of an anchor image, and performing a feature point extraction of a query image. The method also includes clustering feature points of the anchor image and feature points of the query image, where clustering includes determining matching feature points, determining outlier feature points, and excluding outlier feature points. Whether the anchor image is similar to the query image is determined based on a distance between the feature points of the anchor image and the feature points of the query image. If the anchor image is similar to the query image, possible tampered areas of the query image based on the outlier feature points are identified.