CMOS Image Sensor RTS Noise Correction via Synthetic Dark-Time Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CMOS image sensors suffer from random telegraph signal (RTS) noise, which causes defective images due to fluctuations in signal output levels, especially noticeable in miniaturized sensors, leading to reduced image quality and accuracy in noise detection.
Innovation Solution
A photoelectric conversion device and method that includes a generation circuit and a controller to generate a dark-time image signal equivalent to an image signal produced without external light exposure, allowing for accurate noise detection and correction even when the sensor is exposed to external light, using a pseudo light shielding mode that simulates dark-time conditions without physically blocking the light source.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the light source is physically turned off during each scan operation to detect dark-time signals, then noise detection accuracy is improved, but productivity decreases due to repeated start-stop operations
Solution Approach 1:
The patent creates a copy of the dark-time signal characteristics by generating a synthetic dark-time image signal through signal processing. Instead of physically turning off the light source to obtain dark-time signals, the system captures a reference dark-time signal once, stores it, and then generates synthetic dark-time signals by subtracting the reference signal from regular image signals. This copying approach maintains noise detection accuracy while eliminating the need to stop scan operations.
Solution Approach 2:
The patent performs preliminary action by capturing and storing a reference dark-time signal before regular scanning operations begin. This reference signal is obtained when the light source is off, and then it is used to generate synthetic dark-time signals during subsequent scanning operations. By performing the dark-time signal acquisition in advance, the system avoids repeated start-stop operations while maintaining the ability to detect RTS noise accurately.
2Productivity
If the light source remains on during scanning to maintain continuous operation, then productivity is improved, but accurate noise detection becomes difficult due to inability to capture true dark-time signals
Solution Approach 1:
The system copies the essential characteristics of dark-time signals by generating synthetic dark-time image signals through mathematical processing. The synthetic signal is created by subtracting a reference dark-time signal from regular image signals, preserving the RTS noise characteristics needed for accurate detection while allowing the light source to remain continuously on.
Solution Approach 2:
The patent replaces the mechanical approach of physically turning off the light source with a signal processing approach. Instead of using mechanical control to create dark-time conditions, the system uses electronic signal manipulation to generate synthetic dark-time signals, thereby maintaining continuous light source operation while achieving accurate noise detection.
3Volume of moving object
If miniaturization of CMOS sensors is pursued to reduce device size, then device compactness is improved, but RTS noise becomes more noticeable leading to reduced image quality
Solution Approach 1:
The patent extracts and isolates the RTS noise component from the image signal by comparing regular image signals with synthetic dark-time signals. By subtracting the synthetic dark-time signal (which contains RTS noise) from the regular image signal, the system separates the RTS noise component, allowing for its identification and correction while preserving the overall image quality despite miniaturization.
Solution Approach 2:
The system implements feedback by continuously monitoring for RTS noise occurrences and dynamically correcting affected pixels. When RTS noise is detected in real-time through comparison with synthetic dark-time signals, the system identifies the noisy pixels and replaces them with corrected values, thereby maintaining image quality in miniaturized sensors through active noise management.
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 enables high-accuracy noise detection and correction of RTS noise, improving image quality by reducing the impact of RTS noise and increasing productivity by eliminating the need for physically turning off the light source during each scan operation.
Implementation Method 1
a photoelectric conversion element 21 to receive the light reflected by the document
Data Source
AI summary
A photoelectric conversion device includes a generation circuit and a controller. The generation circuit generates an image signal according to an intensity of light being input. The controller controls the generation circuit to generate a dark-time image signal equivalent to an image signal generated by the generation circuit without exposure to external light.