Hierarchical Parallax Calculation for Texture-Less Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parallax calculation techniques using pattern projection for distance measurement face challenges in increasing processing speed and reducing load, particularly in regions with little texture, leading to decreased accuracy.
Innovation Solution
The implementation of hierarchical processing, where a reduced-size image is generated by thinning out captured images at certain pixel intervals, and search ranges are limited based on parallax amounts from lower hierarchies, to reduce calculation load and increase processing speed while maintaining accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pattern light projection is used for distance measurement in regions with little texture, then distance measurement accuracy is improved, but processing load increases and processing speed decreases
Solution Approach 1:
The patent divides the image processing into multiple hierarchies, where the image is first down-sampled to create a coarse-resolution version. Parallax calculation is performed on this reduced image first, then the results are used to guide fine-resolution parallax calculation only in necessary regions. This segmentation of processing resolution allows accurate distance measurement while reducing overall computational load.
Solution Approach 2:
The patent performs preliminary parallax calculation on a down-sampled image before performing detailed parallax calculation on the full-resolution image. The preliminary results from the reduced image are used to determine search ranges and initialize correlation calculations for the fine-resolution image, avoiding redundant computations and improving processing efficiency.
2Measurement precision
If full-resolution image processing is performed for accurate parallax calculation, then distance measurement accuracy is improved, but processing time increases
Solution Approach 1:
The patent introduces a hierarchical dimension to the processing approach, operating at multiple resolution levels rather than a single full-resolution level. By performing initial calculations at lower resolutions and progressively refining results at higher resolutions, the system achieves accurate parallax measurement while significantly reducing total processing time compared to full-resolution processing alone.
Solution Approach 2:
The patent performs preliminary parallax calculation on a down-sampled image before performing detailed parallax calculation on the full-resolution image. The preliminary results from the reduced image are used to determine search ranges and initialize correlation calculations for the fine-resolution image, avoiding redundant computations and improving processing efficiency.
3Measurement precision
If search range is expanded to cover all possible parallax values, then measurement accuracy is improved, but calculation load increases significantly
Solution Approach 1:
The patent performs preliminary parallax calculation on a down-sampled image to obtain approximate parallax values. These preliminary results are then used to define constrained search ranges for the fine-resolution correlation calculation, ensuring that the search covers all possible parallax values while focusing computational effort on the most likely regions, thus reducing overall calculation load.
Solution Approach 2:
The patent applies different processing strategies to different regions of the image based on local characteristics. In regions where accurate parallax measurement is critical, full-resolution processing with expanded search ranges is performed. In other regions, the down-sampled image processing suffices. This local adaptation of processing quality reduces overall calculation load while maintaining necessary measurement accuracy.
Data Source
AI summary
An apparatus comprises: an acquisition unit that captures an image of a measurement target and acquires an original image pair having a parallax; a reduction unit that reduces a size of the original image pair; and a calculator that calculates, from an image pair obtained by the reduction unit, a parallax map using predetermined search ranges and search windows in the respective regions. In order to obtain a parallax map that has a predetermined number of pieces of data, the calculator calculates parallax maps in a plurality of hierarchies in which a hierarchy where a parallax map is calculated using an image pair with a lowest magnification is set as a lowest hierarchy, and the predetermined search ranges for calculating the parallax maps in the respective hierarchies are determined based on a parallax map in an immediately lower hierarchy.