Dynamic Reference Range Image Generation for Adaptive Background Subtraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing range imaging systems struggle with dynamic background subtraction in environments where the background is not constant, leading to inaccurate output, especially in freely accessible areas where objects and lighting conditions change.
Innovation Solution
A method for generating a dynamic reference range image by updating pixels based on measured range values, where values are updated if they remain constant for specific time periods, and invalid values are marked and handled separately to reduce noise and measurement errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a constant reference image is used for background subtraction, then the system is simple to implement, but the system fails when the background changes over time
Solution Approach 1:
The patent implements a dynamic reference image that automatically updates based on temporal consistency of measured range values. The reference image transitions from static to adaptive, updating when pixels maintain constant values across multiple frames, allowing the system to track background changes while maintaining operational simplicity
Solution Approach 2:
The reference image updates itself automatically by comparing current measured range values with stored reference values over time. The system self-adjusts to background changes without external intervention, using the temporal stability of range measurements to autonomously determine when updates are appropriate
2Reliability
If the reference image is updated frequently to track background changes, then the system adapts to dynamic environments, but the system becomes sensitive to noise and measurement errors
Solution Approach 1:
The patent requires that measured range values remain constant for a predetermined number of frames before triggering a reference image update. This preliminary validation step filters out transient noise and measurement errors, ensuring that only stable, reliable changes are incorporated into the reference image
Solution Approach 2:
The system continuously compares current measured range values with the reference image and uses this feedback to determine update timing. By monitoring temporal consistency across multiple frames, the feedback mechanism distinguishes between genuine background changes and random measurement variations, updating only when confidence is sufficient
3Measurement precision
If the system waits for extended periods to confirm background changes, then noise is reduced, but the system responds slowly to actual background changes
Solution Approach 1:
The patent applies different temporal requirements to different spatial locations based on local characteristics. The update threshold and frame count requirements can be adjusted per pixel or region, allowing faster updates in stable areas while maintaining stricter validation in noisy regions, optimizing both speed and precision locally
4Speed
If the system updates the reference image quickly when objects leave the scene, then the system adapts rapidly to changing backgrounds, but the system may incorrectly classify moving objects as background
Solution Approach 1:
The patent requires that range values remain constant for a predetermined number of frames before updating the reference image, even when objects appear to have left the scene. This preliminary confirmation period ensures that transient objects are not mistakenly incorporated into the background, maintaining object detection accuracy while still enabling rapid adaptation to genuine background changes
Data Source
AI summary
A dynamic reference range image generation method comprises providing a reference range image, to be dynamically updated, composed of pixels, each of which contains a reference range value. An acquired range image is provided, the pixels of which contain each a measured range value, the measured range values being updated at a predetermined rate. Pixels of the acquired range image containing an invalid measured range value are accordingly marked. The measured range value of each pixel of the acquired range image not marked as containing an invalid measured range value is compared with the reference range value of the corresponding pixel of the reference range image. The reference range value of that pixel of the reference range image is updated e.g. to the measured range value or to an average of the measured range value and one or more prior measured range values if a) the measured range value is considered less than the reference range value and has remained substantially constant for a first time period, or if b) the measured range value is considered greater than the reference range value and has remained substantially constant for a second time period smaller than the first time period. If neither of conditions a) and b) is fulfilled, the reference range value is kept substantially constant instead.