Hierarchical Stereo Matching Parallax Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In hierarchical stereo matching for distance measurement, when a higher-level parallax is determined to be invalid, all corresponding lower-level parallaxes are often invalidated, leading to a significant number of invalid parallaxes in the final high-resolution parallax image.
Innovation Solution
The image processing device performs a first layer stereo matching to obtain a first parallax and a second layer stereo matching to obtain a second parallax, expanding the search range when the image block with the highest similarity is found at the end of the initial search range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If hierarchical search is used to reduce computation amount and time, then computation efficiency is improved, but the number of invalid parallaxes increases
Solution Approach 1:
The patent performs preliminary coarse parallax calculation at low resolution before fine parallax calculation at high resolution. By pre-establishing a rough parallax value through low-resolution processing, the search range for high-resolution calculation is narrowed, reducing computation time while maintaining accuracy. This preliminary action prevents the need to invalidate all lower-level parallaxes when higher-level parallax is initially deemed invalid.
Solution Approach 2:
The patent applies different processing qualities to different regions: low-resolution processing for broad area coverage and high-resolution processing for localized detailed analysis. By dividing the image into multiple regions and applying hierarchical search selectively, the system maintains high parallax calculation validity in critical areas while reducing overall computation burden.
2Loss of time
If search range is limited to reduce computation time, then processing speed is improved, but parallax accuracy deteriorates
Solution Approach 1:
The patent segments the parallax calculation process into multiple resolution levels. Coarse parallax is calculated first at low resolution with a broader search range, then fine parallax is calculated at high resolution with a narrowed search range centered on the coarse result. This segmentation allows the system to use different search ranges at different stages, improving both speed and accuracy.
Solution Approach 2:
The patent adds a resolution dimension to the parallax calculation process by performing calculations at multiple resolution levels rather than a single resolution. This dimensional approach allows the system to efficiently navigate the trade-off between search range and accuracy by transitioning from low-resolution broad search to high-resolution focused search.
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
This approach increases the possibility of parallax calculation by avoiding the invalidation of all lower-level parallaxes when a higher-level parallax is invalid, allowing for more accurate distance information to be obtained without increasing computation time.
Implementation Method 1
A position measurement device that measures a distance to an object existing in a camera's field of view based on stereo images captured by a stereo camera, which includes a pair of cameras, using the principle of triangulation
Implementation Method 2
block matching, which is performed to search for an image region consisting of a plurality of pixels in one image, the image region having the lowest dissimilarity in the other image
Data Source
AI summary
The invention is directed to an image processing device performs a first layer matching processing for obtaining a first parallax in units of an image region of a first size by specifying an image block having the high similarity to a first image block provided in one of two images from a first search range based on a first predetermined search width in the other image and a second layer matching processing for obtaining a second parallax in units of an image region of a second size by specifying an image block having the high similarity to a second image block provided in one image of the two images from a second search range based on the first parallax and a predetermined second search width in the other image, and expands the first search width or the second search range and performs a search when an image block having the high similarity to the first image block exists at an end of the first search range, in the first layer stereo matching processing.