T-O-F Depth Imaging Phase Correction Electronic Rolling Shutter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Time-of-Flight (T-O-F) depth imaging systems using global demodulators and Electronic Rolling Shutter (ERS) image sensors face inefficiencies in pixel usage and limited frame rates, leading to inaccurate depth images due to sequential capture of light intensity samples, which hampers the ability to accurately image moving objects.
Innovation Solution
A T-O-F depth imaging device employing an ERS scheme with a modulated light source and optical shutter, where the transmission of light and shutter operation change phase relative to each other, allowing for faster frame rates and correcting depth values by computing composite data frames from raw data frames at similar phases to reduce blinking artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If sequential capture of light intensity samples is used, then device complexity is reduced, but frame rate is limited and pixel efficiency is poor
Solution Approach 1:
The pixel array is divided into multiple independently controllable groups or rows, allowing simultaneous capture of light intensity samples at different phases across different segments. This segmentation enables parallel processing of multiple phases, dramatically increasing frame rate while maintaining the simplicity of the global demodulator architecture.
Solution Approach 2:
The patent implements dynamic control of pixel integration timing, where different pixels or pixel groups can start and stop their light integration at different times according to the phase being measured. This dynamic timing control allows the system to efficiently capture multiple phases without requiring complex hardware, thereby improving frame rate while keeping device complexity manageable.
2Device complexity
If sequential capture of light intensity samples is used, then device complexity is reduced, but pixel efficiency is poor
Solution Approach 1:
By segmenting the pixel array into multiple groups that can operate simultaneously at different phases, the system ensures that all pixels are actively capturing useful data throughout the integration period. This eliminates the idle time that occurs in sequential capture, thereby improving pixel efficiency without adding complex hardware.
Solution Approach 2:
The patent enables continuous useful action by ensuring that while one group of pixels is capturing data for one phase, other groups are simultaneously capturing data for different phases. This continuous parallel operation maximizes the utilization of all pixels, improving overall pixel efficiency while maintaining a relatively simple device architecture.
3Productivity
If phase changes during light integration occur, then faster frame rates are achieved, but depth accuracy deteriorates due to phase mismatch
Solution Approach 1:
The patent applies preliminary correction by calculating the phase shift that occurs during the light integration period and using this information to adjust the depth calculation. By anticipating and compensating for the phase change before final depth computation, the system maintains high frame rates while preserving depth accuracy.
Solution Approach 2:
The system uses feedback by measuring the actual phase relationship between the modulated light and the demodulation signal, then using this measured phase information to correct the depth calculations. This feedback mechanism allows the system to maintain accuracy even when phase changes occur during integration, enabling faster frame rates without sacrificing depth precision.
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 enhances pixel efficiency and frame rates, resulting in more accurate depth imaging of moving objects by compensating for phase changes and reducing motion-induced depth oscillation artifacts, thereby improving the overall performance of T-O-F depth imaging systems.
Implementation Method 1
Each individual light sensor, also known as a pixel, responds to light directed onto it by the optical component, the light that ultimately comes from a certain narrow solid angle within the field of view (FOV) of the optical component. This narrow solid angle can be called the acceptance angle or FOV of the pixel. The FOV of an imaging device is simply a sum of the non-overlapping fields of view of all its active pixels.
Implementation Method 2
A T-O-F depth imaging device employing an ERS scheme with a modulated light source and optical shutter, where the transmission of light and shutter operation change phase relative to each other, allowing for faster frame rates
Implementation Method 3
Time-of-Flight (T-O-F) depth imaging systems using global demodulators and Electronic Rolling Shutter (ERS) image sensors face inefficiencies in pixel usage and limited frame rates
Data Source
AI summary
In embodiments, a T-O-F depth imaging device renders a depth image of an object that has corrected depth values. The device includes a pixel array that uses an Electronic Rolling Shutter scheme. The device also includes a light source that transmits towards the object light that is modulated at an operating frequency, and an optical shutter that opens and closes at the operating frequency. The optical shutter further modulates the light that is reflected from the object, before it reaches the pixel array. The transmission of the light and the operation of the shutter change phase relative to each other while at least one of the pixels is sensing the light it receives, which permits a faster frame rate. Depth is determined by the amount of light sensed by the pixels, and a correction is computed to compensate for the changing phase.