Parallax Search Window Tuning for Image-Height Distance Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distance measurement devices using the pupil division phase difference method face accuracy variations due to image height, as the baseline length changes with vignetting, affecting the precision of distance calculations.
Innovation Solution
A distance measurement device that sets the search window based on optical characteristic values corresponding to the centroidal gap between luminous fluxes on the exit pupil, adjusting the window size according to image height to stabilize baseline length and reduce parallax calculation errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the pupil division phase difference method is used with one optical system, then the device complexity is reduced, but the measurement precision varies with image height due to vignetting causing baseline length variation
Solution Approach 1:
The patent applies local quality by setting different search window sizes according to image height regions. Specifically, the search window size is adjusted based on the baseline length variation caused by vignetting at different image heights, with smaller windows for peripheral regions where baseline variation is more significant. This localized adaptation maintains measurement precision across different image heights while preserving the simplicity of the single optical system configuration.
2Device complexity
If the search window size is fixed, then the calculation process is simplified, but the measurement precision varies across different image heights
Solution Approach 1:
The patent implements dynamics by making the search window size variable rather than fixed. The search window size is dynamically adjusted according to the image height and corresponding baseline length. This dynamic adaptation allows the system to maintain optimal measurement precision across different image heights while adding only minimal complexity to the parallax calculation process through automated window size selection.
3Measurement precision
If distortion correction is applied to maintain image quality, then the image processing complexity increases, but the baseline length variation with image height persists
Solution Approach 1:
The patent applies parameter changes by modifying the search window size parameter based on image height and baseline length characteristics. Rather than attempting to correct all optical distortions, the invention changes the search window parameter to adapt to the varying baseline conditions at different image heights. This approach addresses the measurement precision issue through parameter adaptation rather than through complex distortion correction processing.
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 reduces variation in distance measurement accuracy across different image heights, enhancing the precision of parallax calculations and overall distance measurement.
Implementation Method 1
Each pixel of an image capturing element is constituted to have divided photoelectric conversion portions while sharing a microlens, and each of the photoelectric conversion portions receives one of luminous fluxes which have passed through different pupil regions in one optical system, thereby acquiring a set of images having parallax therebetween.
Implementation Method 2
calculate an amount of parallax of a set of images having parallax therebetween and captured from different visual points
Data Source
Figure 1
Figure 2A~2C
Figure 3A~3B
AI summary
A distance measurement device includes an image capturing unit configured to be capable of capturing images of first image information and second image information having parallax therebetween from one optical system, and a parallax calculation unit configured to extract image information within a search window from each of the first image information and the second image information and calculate an amount of parallax using a correlation between image information of the first image information within the search window and image information of the second image information within the search window which have been extracted. The parallax calculation unit sets the search window on the basis of an optical characteristic value of an image height for performing parallax calculation. The optical characteristic value corresponds a centroidal gap between a centroid of a first luminous flux forming the first image information and a centroid of a second luminous flux forming the second image information on an exit pupil of the optical system.