Virtual-Pixel Tessellation for CMOS TOF Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solid-state imaging devices face challenges in achieving high resolution for range imaging while maintaining precision, particularly with the decrease in pixel sizes leading to reduced range-measurement precision and complexity in quadruple-tap pixel configurations.
Innovation Solution
A solid-state imaging device with virtual-pixel units arranged in a polygonal tessellation pattern, incorporating multiple charge detectors and transfer-control elements, enabling both summed-readout and cyclic-readout modes to enhance signal processing and precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of CMOS TOF pixels is increased to achieve higher resolution range imaging, then the imaging resolution is improved, but the range-measurement precision deteriorates due to drops in sensibility and number of saturated electrons with reduced pixel sizes
Solution Approach 1:
The patent applies binning by combining signals from multiple adjacent TOF pixels (e.g., 2x2 or larger groups) to form super-pixels. This merging approach increases the effective pixel size and electron capacity while maintaining high imaging resolution through the array of combined pixels, thereby resolving the contradiction between individual pixel precision and overall system resolution.
Solution Approach 2:
The patent segments the pixel array into multiple tap groups (e.g., quadruple-tap configuration with 4 separate readout paths) that can be independently controlled and read out. This segmentation allows selective binning of adjacent pixels with the same tap number while maintaining the ability to perform precise phase measurements through multiple taps, thus preserving both precision and resolution.
2Measurement precision
If binning is applied to increase pixel capacity and maintain precision, then range-measurement precision is improved, but the configuration complexity increases significantly making quadruple-tap binning difficult to achieve
Solution Approach 1:
The patent implements a universal binning architecture where the same tap structure (e.g., quadruple-tap) is applied across all pixel groups, with standardized readout circuits and control logic. This multi-functional design allows the system to perform both individual pixel readout and binning operations using the same hardware framework, reducing configuration complexity while maintaining precision.
Solution Approach 2:
The patent pre-organizes adjacent pixels into fixed binning groups with predetermined tap assignments during the pixel array design phase. This preliminary structuring of pixel groups and their associated readout circuits simplifies the control logic and reduces runtime configuration complexity, making quadruple-tap binning achievable without significant system complexity.
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
The solution allows for both summed-readout and cyclic-readout modes, improving the imaging device's ability to maintain precision and resolution, even with smaller pixel sizes, by effectively managing signal charges and reducing complexity in pixel configurations.
Implementation Method 1
each of the virtual-pixel units encompassing a photoelectric-conversion region
Data Source
AI summary
[Problem] To provide a solid-state imaging device enabling summed-readout mode and cyclic-readout mode.[Solution]A device includes virtual-pixel units implemented by pixels of transfer-route controlling-scheme, the virtual-pixel units having a shape of polygon, the polygons are tessellated. The virtual-pixel unit encompasses a photoelectric-conversion region, charge detectors to which ordinal numbers are labeled, configured to accumulate signal charges transferred from the photoelectric-conversion region, and transfer-control elements configured to control movement of signal charges from the photoelectric-conversion region to one of the charge detectors. N pieces of charge detectors of the same ordinal number are arranged in boundary between the photodiodes and the intersection-shared sites. The intersection-shared site encompasses N pieces of switching elements having first main electrode connected to the charge detectors, and common signal-readout circuit having input terminals connected to second main electrode of each of the switching elements.


