Spatial Coded Slide Image Alignment for 3D Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing three-dimensional scanning systems using structured light stereoscopy face challenges in maintaining alignment of epipolar lines due to lens distortion, particularly in projector lenses, which affects the accuracy of 3D measurements and requires complex corrections or reduced resolution.
Innovation Solution
A method to prepare a spatial coded slide image by obtaining distortion vectors for projector coordinates, aligning the pattern along ideal epipolar lines, and compensating for lens distortion to ensure accurate projection and alignment of coded patterns along epipolar lines, even in the presence of projector lens distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lens distortion is not compensated, then device complexity is reduced, but measurement precision deteriorates due to misalignment of epipolar lines
Solution Approach 1:
The patent applies preliminary action by pre-computing distortion vectors for all projector coordinates before actual 3D scanning. The distortion compensation data is prepared in advance and stored, so that during operation, only simple lookup and application of pre-computed vectors is needed, rather than performing complex distortion calculations in real-time.
Solution Approach 2:
The patent replaces complex mechanical/optical alignment adjustments with a computational approach. Instead of physically adjusting the projector-camera alignment to account for distortion, the system uses software-based distortion vector compensation to achieve precise epipolar line alignment, substituting mechanical precision requirements with computational correction.
2Measurement precision
If distortion vectors are computed and applied for all pixels, then measurement precision is improved, but processing time increases
Solution Approach 1:
The distortion vectors are computed in advance during an initialization phase, separate from the actual 3D scanning operation. This pre-computation allows the time-consuming distortion analysis to be performed once, with results cached for rapid application during subsequent scanning operations, minimizing time loss during actual measurement.
Solution Approach 2:
The system performs self-calibration by automatically computing distortion vectors from the projector's own projection data and the camera's capture data. The system uses its own operational data to generate the compensation parameters it needs, eliminating the need for external calibration equipment or manual intervention.
3Productivity
If epipolar lines are not aligned, then device complexity is reduced, but productivity deteriorates due to reduced measurement accuracy
Solution Approach 1:
The patent replaces complex mechanical alignment systems with a software-based distortion compensation approach. By computing and applying distortion vectors, the system achieves precise epipolar line alignment without requiring complex mechanical adjustment mechanisms or precise physical alignment of the projector-camera setup.
Solution Approach 2:
The patent changes the parameter representation by working with distortion vectors that describe the deviation from ideal epipolar geometry. By transforming the alignment problem from a geometric positioning problem to a parameter correction problem, the system can achieve high precision through computational parameter adjustment rather than physical reconfiguration.
Data Source
Figure 1A~1C
Figure 2
Figure 3A
AI summary
A method for preparing a spatial coded slide image in which a pattern of the spatial coded slide image is aligned along epipolar lines at an output of a projector in a system for 3D measurement, comprising: obtaining distortion vectors for projector coordinates, each vector representing a distortion from predicted coordinates caused by the projector; retrieving an ideal pattern image which is an ideal image of the spatial coded pattern aligned on ideal epipolar lines; creating a real slide image by, for each real pixel coordinates of the real slide image, retrieving a current distortion vector; removing distortion from the real pixel coordinates using the current distortion vector to obtain ideal pixel coordinates in the ideal pattern image; extracting a pixel value at the ideal pixel coordinates in the ideal pattern image; copying the pixel value at the real pixel coordinates in the real slide image.