3D Reflection Detection for Privacy Masking in Image Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques struggle to effectively identify and mask reflections of objects from reflective surfaces in surveillance images, leading to potential breaches of privacy and incomplete data anonymization.
Innovation Solution
A method involving the creation of a three-dimensional representation of a scene from a captured image frame, tracing optical rays to detect reflective surfaces that match the color of the object, and applying image processing to mask these reflections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If masking techniques are applied to protect privacy in video surveillance, then personal privacy is protected, but reflective surfaces may still reveal masked objects through reflections
Solution Approach 1:
The system performs preliminary ray tracing calculations before applying masks to identify all pixels that will display reflections of masked objects. By pre-calculating reflection paths using the three-dimensional scene model and comparing color values, the system prepares a complete set of pixels requiring masking, ensuring no reflective information leaks through overlooked areas.
Solution Approach 2:
The patent introduces a three-dimensional scene model as an intermediary between the raw video feed and the final masked output. This model serves as a mediator that enables accurate ray tracing calculations to identify reflective surfaces, allowing the system to precisely determine which pixels need masking without affecting the quality of non-reflective areas.
2Reliability
If dynamic masking is applied to move with objects, then privacy protection follows the object, but computational complexity increases significantly
Solution Approach 1:
The system performs preliminary action by pre-calculating reflection paths using ray tracing in a three-dimensional scene model before applying dynamic masks. This allows the system to identify all reflective pixels in advance, simplifying the actual masking application process while maintaining accuracy.
Solution Approach 2:
The patent creates a three-dimensional copy or model of the scene that mirrors the physical environment. This virtual model allows ray tracing calculations to be performed efficiently without directly processing the complex real-time video data, reducing computational burden while maintaining accuracy.
3Measurement precision
If ray tracing is used to identify reflective surfaces, then detection accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The system creates a three-dimensional model copy of the scene that can be used for ray tracing calculations. This virtual model allows accurate reflection path computation without requiring intensive processing of actual video frames, significantly reducing processing time while maintaining detection accuracy.
Solution Approach 2:
The patent performs ray tracing calculations in advance using the three-dimensional scene model to identify all reflective pixels before the actual masking process. This preliminary identification step allows the system to prepare a complete mask map, avoiding repeated calculations during real-time processing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances data anonymization by accurately identifying and masking reflections, ensuring privacy protection while maintaining the ability to analyze movements and activities in surveillance footage.
Implementation Method 1
tracing optical rays in the three-dimensional representation of the scene from the identified object to the camera via reflecting surfaces in the scene
Data Source
AI summary
An image-processing device generates a three-dimensional model of a background scene of the image frame based on three-dimensional information about the background scene. The image-processing device defines a three-dimensional bounding box of the object in the three-dimensional model. The image-processing device defines a centre coordinate in the three-dimensional model and a colour value of surface elements of the three-dimensional bounding box. The image-processing device determines a three-dimensional coordinate of a surface in the three-dimensional model which reflects light from a surface element into the camera, by tracing rays from the centre coordinate and based on a normal of the surface. The image-processing device further identifies a first pixel in the image frame corresponding to the three-dimensional coordinate and detects the reflection of the object.