Photon Counting with Pixel Noise Maps for Accurate CMOS Readout
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Photon counting using CMOS image sensors is hindered by reading noise, which broadens the probability distribution of photoelectrons and reduces counting accuracy in pixels with high noise variability.
Innovation Solution
A photon counting device and method that derive a confirmed photon count value by considering both the photon number distribution and reading noise distribution, using a first and second probability calculation to minimize the influence of reading noise, and employing weighted averages and noise maps to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If reading noise is reduced in the amplifier, then measurement precision of photon number is improved, but device complexity increases due to additional noise correction processing
Solution Approach 1:
The patent performs preliminary characterization of reading noise for each pixel by capturing multiple dark frames (without light input) and calculating the standard deviation of signal values. This noise profile is stored and used in subsequent photon counting measurements to correct for pixel-specific noise variations, thereby improving measurement precision without requiring hardware modifications.
Solution Approach 2:
The patent implements a feedback mechanism where the measured reading noise from dark frames is used to adjust and correct the photon number calculation in actual measurement frames. The system continuously refines its noise model and applies corrections to compensate for reading noise, improving counting accuracy through iterative noise characterization and correction.
2Reliability
If reading noise is reduced in the amplifier, then reliability of photon counting is improved, but manufacturing precision requirements increase
Solution Approach 1:
The patent applies local quality by characterizing and correcting reading noise individually for each pixel rather than assuming uniform noise across the sensor array. Each pixel's noise profile is measured separately using dark frames and applied as a specific correction factor, allowing the system to handle manufacturing variations in amplifier noise without requiring tight manufacturing tolerances.
Solution Approach 2:
The patent changes the parameter approach from attempting to manufacture amplifiers with uniform noise characteristics to measuring and correcting the actual noise parameters of each pixel. By capturing multiple dark frames and calculating statistical parameters (mean and standard deviation) for each pixel, the system adapts to manufacturing variations through parameter-based correction rather than relying on manufacturing 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
The solution improves photon counting accuracy by reducing the impact of reading noise, ensuring precise photon number determination even in pixels with varying noise levels.
Implementation Method 1
a photodiode converting input light to charge
Data Source
AI summary
A photon counting device includes: a plurality of pixels each including a photoelectric conversion element and an amplifier amplifying charge to which input light is converted by the photoelectric conversion element and converting the amplified charge to a voltage; an A/D converter converting a voltage output from the amplifier of each of the plurality of pixels to a digital value; a first derivation unit configured to derive a provisional value of photon number in each of the plurality of pixels based on the digital value; and a second derivation unit configured to derive a confirmed value of photon number in a target pixel based on a first probability associated with a photon number distribution and a second probability associated with reading noise.