Stereo Matching Apparatus Vertical Disparity Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Stereo matching in image processing systems faces challenges due to vertical and rotational differences between camera positions, especially when caused by dynamic factors like vibration, leading to increased processing time and potential mismatches.
Innovation Solution
An image processing apparatus that includes a first cost calculating section for horizontal similarity search, a second cost calculating section for vertical similarity search, an amount-of-difference buffer to record vertical position differences, and a search position setting section to determine optimal search positions, allowing for efficient disparity calculation even with dynamic camera differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the similarity degree search range is spread in the vertical direction to handle vertical differences between cameras, then matching accuracy is improved, but the number of functional units required for similarity degree calculation increases and processing time is lengthened
Solution Approach 1:
The patent divides the similarity degree calculation into two separate stages: horizontal similarity calculation (first cost calculating section) and vertical similarity calculation (second cost calculating section). This segmentation allows the system to handle vertical differences without requiring all functional units to operate simultaneously, thus improving matching accuracy while controlling device complexity.
Solution Approach 2:
The patent performs horizontal similarity calculation first to identify candidate matching positions, then uses these results to guide the subsequent vertical similarity calculation. This preliminary action reduces the search space for vertical comparison, avoiding the need to calculate vertical similarity for all possible positions and thereby reducing the number of functional units required.
2Measurement precision
If the similarity degree search range is spread in the vertical direction to handle vertical differences between cameras, then matching accuracy is improved, but processing time is lengthened
Solution Approach 1:
By separating horizontal and vertical similarity calculations into distinct stages, the patent enables parallel processing within each stage and reduces the overall computational burden. The first cost calculating section processes horizontal similarities, and the second cost calculating section processes vertical similarities only at identified candidate positions, significantly reducing total processing time while maintaining matching accuracy.
Solution Approach 2:
The horizontal similarity calculation serves as a preliminary filtering step that identifies promising candidate positions before vertical similarity calculation begins. This preliminary action eliminates the need for exhaustive vertical comparison across the entire image, thereby reducing processing time while preserving matching accuracy through targeted vertical verification.
3Device complexity
If two cameras are arranged horizontally for stereo matching, then disparity calculation is simplified, but vertical and rotational differences between cameras cannot be compensated
Solution Approach 1:
The patent introduces dynamic adjustment of the search range in the vertical direction based on the amount of difference information stored in the amount-of-difference buffer. This dynamic approach allows the system to adapt to varying vertical and rotational differences between cameras while maintaining the simplified horizontal arrangement, thereby improving matching reliability without significantly increasing disparity calculation complexity.
Solution Approach 2:
The patent introduces an intermediary mechanism (the amount-of-difference buffer and search position setting section) that mediates between the fixed horizontal camera arrangement and the variable vertical misalignment. This intermediary dynamically adjusts the search parameters to compensate for vertical and rotational differences, maintaining matching reliability while preserving the simplicity of the horizontal camera configuration.
Data Source
AI summary
An image processing apparatus includes: a first cost calculating section configured to calculate similarity degrees between a first pixel in a first image and plural second pixels located in a horizontal-direction first range in a second image, and extract a second pixel with a highest similarity degree; a second cost calculating section configured to calculate a similarity degree between plural third pixels in a vertical-direction second range including the extracted second pixel and the first pixel and extract a third pixel with a highest similarity degree; an amount-of-difference buffer recording difference between a vertical-position of the extracted third pixel and the first pixel as an amount of difference; and a second position setting section configured to determine positions of the second pixels for which a similarity degree to the first pixel for which a disparity is calculated next is calculated.