Stereo Matching Feature Point Estimation for Processing Load Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing range image acquisition techniques using stereo matching face high processing loads due to dense distance value calculations, which can lead to increased noise and reduced efficiency in detecting feature points of objects in three-dimensional space.
Innovation Solution
The proposed solution involves an information processing apparatus that performs stereo matching on a limited number of candidate points and their surrounding points within a predetermined distance range, reducing the calculation cost by selectively setting candidate points and determining noise based on the distribution of distance values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stereo matching is performed on all pixels to generate dense range image, then measurement precision is improved, but processing time increases significantly
Solution Approach 1:
The patent segments the image into multiple local areas and processes each area independently. Instead of performing stereo matching on all pixels globally, the system divides the workspace into local regions and conducts matching operations within each region, reducing overall processing time while maintaining measurement precision through localized statistical analysis.
Solution Approach 2:
The patent applies partial action by performing stereo matching only on a subset of pixels rather than all pixels. By selecting representative pixels and performing matching on limited areas, the system achieves sufficient measurement precision without the excessive processing time required for complete pixel-by-pixel matching.
2Reliability
If noise determination is performed using statistical calculation on multiple distance values, then reliability is improved, but processing load increases
Solution Approach 1:
The patent segments the image into multiple local areas and performs statistical noise determination independently in each area. This segmentation allows the system to apply reliable statistical calculations on multiple distance values within manageable local regions, maintaining noise determination accuracy while avoiding the excessive processing load of global statistical analysis.
Solution Approach 2:
The patent applies local quality by performing noise determination with statistical calculations only in local areas where needed, rather than uniformly across the entire image. This approach maintains high reliability in critical regions while improving overall processing efficiency by limiting intensive calculations to specific local zones.
Data Source
AI summary
An information processing apparatus acquires a stereo image, performs matching of feature points of a first number smaller than the number of pixels in a first image to estimate three-dimensional positions of the feature points with respect to a stereo camera, sets the feature point determined to be acquired from a space set in a field of view of an imaging unit as a target point, sets surrounding points of a second number greater than the number of the feature points for which the three-dimensional positions are estimated in an image area within a predetermined distance range from the target point in the first image, and determines whether the target point is the feature point indicating a feature of an object existing in the space based on differences between the three-dimensional positions of the surrounding points and the target point with respect to the stereo camera.