Coded Light 3D Camera Frame Rate via Pattern Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current coded light 3D geometry acquisition systems face limitations in frame rate due to the need to project additional dark and bright patterns, which are affected by object motion and reflectivity variations, leading to incorrect binarization and reduced frame rates.
Innovation Solution
The system eliminates the need to project explicit dark and bright patterns by using a one-dimensional array of optical sub-systems with laser diodes and diffraction optical elements, where binary code patterns are created through electronic switching, and dark and bright pixel values are reconstructed from the minimum and maximum pixel values over a code sequence, or by projecting alternating normal and inverted patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If explicit dark and bright patterns are projected for binarization, then binarization accuracy is improved, but frame rate decreases
Solution Approach 1:
The patent extracts and removes the explicit dark and bright reference patterns from the projection sequence. Instead of projecting these patterns separately, the system reconstructs reference values by selecting minimum and maximum pixel intensities from the code pattern sequence itself, thereby eliminating the overhead of additional reference pattern projections and improving frame rate.
Solution Approach 2:
The code patterns serve multiple functions: they encode depth information for 3D reconstruction and simultaneously provide the intensity range data needed for binarization. By using the same patterns for both depth encoding and reference intensity extraction, the system eliminates the need for separate reference patterns while maintaining binarization accuracy.
2Stability of the object's composition
If dark and bright patterns are updated at low frequency, then binarization stability is improved, but motion compensation capability deteriorates
Solution Approach 1:
The patent makes the reference intensity values dynamic by extracting them from each code pattern pair rather than using fixed, periodically updated reference patterns. This allows the reference values to adapt to changing scene conditions and object motion in real-time, improving motion compensation capability while maintaining binarization stability through consistent extraction methodology.
3Measurement precision
If more code patterns are projected for higher resolution, then 3D reconstruction precision is improved, but frame rate decreases
Solution Approach 1:
The system uses the code pattern sequence itself to provide the reference intensity information needed for binarization, rather than requiring separate reference patterns. This self-service approach eliminates the overhead of additional pattern projections, allowing more code patterns to be used for higher resolution without the same penalty to frame rate.
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 frame rate by reducing the number of patterns required for 3D frame acquisition, allowing for higher frame rates and improved motion compensation in coded light systems.
Implementation Method 1
The proposed design involves a fixed on-dimensional array of optical sub-systems, consisting of light sources (e.g. laser diodes or vertical cavity surface emitting diode lasers (VCSEL))
Implementation Method 2
coupled with an array of masks (diffraction optical element (DOE) or simple optical filters)
Implementation Method 3
Coded light is a method for three-dimensional (3D) geometry acquisition, in which the object of interest is illuminated by a time-multiplexed sequence of patterns
Data Source
AI summary
In a time-multiplexed structured (coded) light camera, the coded light patterns are formed by means of an array of laser sources, and binarizations of the light patterns are performed by computing the dark and bright frames from the code patterns themselves.