Video Anonymization for Stationary Person Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video management systems and CCTV systems fail to effectively anonymize stationary individuals in surveillance images, as they become part of the background image and are not obscured, despite the need for privacy protection.
Innovation Solution
A video processing controller with a Background Subtraction Module, Skin Detection Module, and Anonymization Module processes images to identify and obscure both moving and stationary individuals by using pixelation, blurring, or other obscuration methods, ensuring privacy protection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If background subtraction models are used to detect moving objects, then moving objects can be extracted and anonymized, but stationary persons become part of the background image and are not anonymized
Solution Approach 1:
The system divides the detection task into two separate modules: one for detecting moving objects through background subtraction, and another for detecting stationary persons with skin tones. This segmentation allows each module to specialize in its target type, resolving the contradiction between detecting moving objects and covering stationary persons.
Solution Approach 2:
The system creates a universal anonymization approach that handles both moving and stationary persons through multiple detection mechanisms. By combining background subtraction with skin tone detection, the system achieves multi-functionality in detecting different types of persons, ensuring comprehensive privacy protection.
2Loss of information
If skin-based privacy filters are applied, then human skin areas can be identified and obscured, but the system may not effectively handle stationary individuals who blend into the background
Solution Approach 1:
The system introduces skin tone detection as an intermediary mechanism that works in parallel with background subtraction. This intermediary detection method specifically targets human skin areas regardless of movement status, ensuring that stationary persons who blend into the background are still reliably detected and anonymized.
3Productivity
If traditional background subtraction is used, then moving objects can be detected, but the complexity of the system increases when attempting to also detect stationary objects
Solution Approach 1:
The system segments the detection functionality into distinct modules: background subtraction for moving objects and skin tone detection for stationary persons. This modular segmentation maintains processing efficiency by allowing each module to operate independently with optimized algorithms, while avoiding the complexity of a single unified system attempting to handle all detection scenarios.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A video management system may identify moving objects in a scene, and may obscure the moving object. Additionally, the video management system may identify parts of the scene that are not moving, and may identify pixels having a color falling within a range which may be associated with human skin, and may obscure such pixels. The video management system may present an image to a monitor including the obscured moving object and the obscured pixels having the color falling within the range associated with human skin.