Tamper Detection Using Edge Weight and Intensity Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capturing devices face challenges in automatically detecting tampering, such as camera movement, changes in lighting, and obstructions, without triggering false alarms from benign events like wind or large objects in the scene.
Innovation Solution
A method that assesses tampering by calculating a weighted score based on instantaneous image intensity changes, foreground mask area, and edge overlap between current and background frames, using edge detection and illumination metrics to determine if the camera has been tampered with.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If simple image difference methods are used to detect tampering, then detection speed is improved, but false alarms increase due to inability to distinguish benign events from actual tampering
Solution Approach 1:
The patent segments the image analysis into multiple independent metrics: intensity change detection, foreground mask area calculation, and edge overlap measurement. Each metric handles a specific aspect of tamper detection, allowing the system to process information in parallel and make more accurate distinctions between benign events and actual tampering without sacrificing speed.
Solution Approach 2:
The patent applies different analysis methods to different regions of the image. Edge detection focuses on boundary regions to identify camera movement, while intensity change analysis examines overall illumination changes. This localized quality approach allows each region to be processed according to its specific characteristics, improving both speed and accuracy.
2Measurement precision
If multiple metrics are combined for tamper detection, then detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent merges three distinct detection metrics into a unified tamper detection system: intensity change detection, foreground mask area calculation, and edge overlap measurement. These metrics are combined through a weighted scoring mechanism that integrates their results, achieving high detection accuracy while managing computational complexity through efficient algorithm design.
3Difficulty of detecting and measuring
If the system is sensitive to detect all changes, then detection capability is improved, but false alarms from benign events like wind and large objects increase
Solution Approach 1:
The patent employs feedback mechanisms where the edge overlap metric provides contextual information about the scene structure, and the foreground mask area metric provides feedback about the spatial distribution of changes. This feedback allows the system to distinguish between benign events (where edges remain consistent) and actual tampering (where edge relationships are disrupted), reducing false alarms while maintaining high detection capability.
Data Source
AI summary
A method and apparatus for detecting tampering with an image capturing device. The method includes determining a change has occurred in average intensity if an image, determining a percentage area of a foreground mask, detecting related at least one edge and determining at least one edge weight, utilizing the average intensity, the percentage area of the foreground mask, and the at least one edge weight to determining offset of edge blocks of an image, and utilizing the determining offset of edge blocks to detect tampering with an image capturing device.