High Dynamic Range Pixel Dual-Mode Capacitance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Standard pixels in Time-Of-Flight imaging systems face saturation issues and high noise levels, leading to lost information and image artifacts, and existing methods to increase dynamic range either reduce saturation or noise floor, but not both effectively, while also complicating post-processing for logarithmic pixels.
Innovation Solution
A High Dynamic Range pixel using a dual-mode MOS capacitance that allows non-destructive multiple read-outs with different conversion gains, enabling charge transfer between parasitic and MOS capacitances, and operating in inversion and accumulation modes to achieve high sensitivity and low noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard pixels are used in Time-Of-Flight imaging systems, then the device structure is simple, but pixel saturation occurs and noise levels are high, leading to lost information and image artifacts
Solution Approach 1:
The pixel is segmented into two distinct capacitance elements: a first capacitance element for storing charge in high illumination conditions and a second capacitance element for storing charge in low illumination conditions. This segmentation allows the pixel to handle both bright and dark scenes effectively, preventing saturation and reducing noise without requiring a completely new pixel architecture.
Solution Approach 2:
The pixel dynamically switches between using the first capacitance element and the second capacitance element based on illumination conditions. The conversion gain is dynamically adjusted by selecting which capacitance element to use, allowing the pixel to adapt to varying light levels and maintain optimal performance across different scene brightness levels.
2Duration of action of moving object
If the dynamic range is increased by reducing noise floor, then the saturation level decreases, but if the saturation level is increased, then the noise floor increases
Solution Approach 1:
The invention changes the capacitance parameter by providing two different capacitance values (first and second capacitance elements) instead of using a single fixed capacitance value. This allows the system to select the appropriate capacitance based on illumination conditions, effectively expanding the dynamic range while maintaining signal quality across both high and low light levels.
3Duration of action of moving object
If multiple read-outs with different conversion gains are performed, then the dynamic range is improved, but the charge information is destroyed in standard approaches
Solution Approach 1:
The first and second capacitance elements act as intermediaries that store charge information in different states. The first capacitance element stores charge for high conversion gain read-out, while the second capacitance element stores charge for low conversion gain read-out. This intermediary storage mechanism enables multiple read-outs with different conversion gains without destroying the original charge information.
4Duration of action of moving object
If logarithmic pixels are used to increase dynamic range, then the saturation and noise issues are addressed, but post-processing complexity increases
Solution Approach 1:
Instead of using logarithmic transformation of the signal, the invention changes the physical capacitance parameter to achieve dynamic range extension. By selecting between two fixed capacitance values rather than applying logarithmic processing, the system achieves similar dynamic range benefits while maintaining linear signal characteristics that are easier to process.
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
Enables consistent data acquisition with both low and high conversion gains without destroying charge information, improving dynamic range and reducing noise, particularly beneficial for Time-Of-Flight measurements by allowing multiple correlations to be combined mathematically without calibration.
Implementation Method 1
the photodiode converts the incident light into charges which are accumulated during a given exposure time
Implementation Method 2
A High Dynamic Range pixel using a dual-mode MOS capacitance that allows non-destructive multiple read-outs with different conversion gains, enabling charge transfer between parasitic and MOS capacitances
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The invention relates to a HDR pixel comprising a photo-sensitive element; a detector node connected to the photo-sensitive element; a reset switch connected to the detector node for resetting the detector node to a predetermined voltage; a buffer amplifier having an input connected to the detector node; a selecting transistor operable to select said pixel during a read out process; an intrinsic parasitic capacitance originated from at least one of the photo-sensitive element, the detector node, the reset switch, the buffer amplifier, the selecting transistor and operable to store the minority carriers generated by the photo-sensitive element; characterized in that the pixel further comprises a dual-mode capacitance having an input connected to the detector node and being operable in storing and destoring modes, for storing the generated minority carriers while being in the storing mode and destoring the minority carriers into the parasitic capacitance, while being in the destoring mode.