Image Sensor Pixel Array Sub-pixel Averaging Signal-to-Noise Ratio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CMOS image sensors face challenges in achieving high signal-to-noise ratios and reducing random noise, which affect the quality of images produced.
Innovation Solution
The implementation of a pixel array with multiple sub-pixels that are simultaneously selected by row control signals, allowing for the averaging of sub-pixel signals to generate an analog pixel signal, which is then converted to a digital signal by an analog-to-digital converter, thereby increasing full well capacity and conversion gain.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single pixel structure is used, then device complexity is low, but signal-to-noise ratio is insufficient and random noise is high
Solution Approach 1:
The pixel is divided into multiple sub-pixels (first sub-pixel and second sub-pixel) that can be independently controlled. Each sub-pixel has its own photoelectric conversion element and transfer transistor, allowing separate signal generation and then combination. This segmentation enables the system to achieve higher signal-to-noise ratio by aggregating signals from multiple independent units while maintaining manageable complexity through modular design.
Solution Approach 2:
The patent combines signals from multiple sub-pixels by transferring charges to a common floating diffusion node. The first and second sub-pixels transfer their generated charges to the same accumulation node, where signals are merged and averaged. This merging process improves signal-to-noise ratio through signal aggregation while the shared readout path keeps the overall device complexity controlled.
2Quantity of substance
If multiple sub-pixels are combined, then full well capacity increases, but conversion gain may be reduced
Solution Approach 1:
Each sub-pixel is designed with identical photoelectric conversion elements and transfer transistors, ensuring uniform local characteristics. The floating diffusion node is shared among all sub-pixels, creating a common conversion region with optimized capacitance. This local quality consistency ensures that while full well capacity increases through multi-sub-pixel aggregation, conversion gain is maintained at the shared conversion node through proper capacitance matching.
3Measurement precision
If sub-pixel signals are averaged, then random noise is reduced, but device complexity increases
Solution Approach 1:
The floating diffusion node automatically performs signal averaging through its inherent capacitance properties. When multiple sub-pixels transfer their charges to the shared floating diffusion node, the node naturally integrates and averages the signals based on capacitance division. This self-service averaging mechanism reduces random noise through statistical accumulation without requiring complex external signal processing circuits, thereby minimizing additional device 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
This approach enhances the signal-to-noise ratio and reduces random noise, resulting in improved image quality by increasing both full well capacity and conversion gain simultaneously.
Implementation Method 1
Each of the first and second sub-pixels includes a photodiode and a transfer transistor. The first photodiode and the second photodiode are different from each other in a type of the photodiode.
Data Source
AI summary
An image sensor having improved signal-to-noise ratio and reduced random noise and an image processing system are provided. The image sensor includes a pixel array including a pixel connected to a column line and configured to provide an analog pixel signal to the column line in response to at least one row control signal, and an analog-to-digital converter (ADC) that receives and converts the analog pixel signal into a corresponding digital pixel signal. The pixel includes a group of sub-pixels simultaneously selected by the at least one row control signal, such that each one of the sub-pixels in the group of sub-pixels provides a sub-pixel signal, and the analog pixel signal is an average of the sub-pixel signals provided by the group of sub-pixels.