Image Sensor Noise Compensation via AC Component Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CMOS image sensors face challenges in reducing pixel coupling noise and power noise, which degrade image quality.
Innovation Solution
The image sensor design includes a pixel array with monitoring pixels and active pixels, a binning circuit, an analog-to-digital converter, and a control circuit that extracts an alternating current (AC) component from monitoring signals to generate a compensation signal, which is used to couple with detection signals and reduce noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If monitoring pixels are added to extract and compensate for noise, then noise compensation capability is improved, but device complexity increases
Solution Approach 1:
The pixel array is segmented into active pixels for image capture and separate monitoring pixels for noise measurement. The monitoring pixels are distributed across multiple rows and columns, with each group sharing a common monitor output line. This segmentation allows independent noise measurement without affecting image capture functionality.
Solution Approach 2:
The monitoring pixels serve multiple functions: they measure pixel coupling noise, measure power noise, and their signals are processed through binning circuits and ADCs to generate compensation signals. The same monitoring infrastructure is used for different noise types, making the system multi-functional.
2Device complexity
If multiple monitoring pixels share a common monitor output line, then device complexity is reduced, but signal processing complexity increases
Solution Approach 1:
Multiple monitoring pixels that experience similar noise conditions are merged by connecting them to a common monitor output line. Their signals are then combined through binning circuits, which sum the signals and divide by the number of pixels to produce an average monitoring signal. This merging approach reduces the number of separate signal paths while maintaining noise measurement accuracy.
Solution Approach 2:
The average monitoring signal, which represents noise characteristics, is fed back to the ADC and used to generate compensation signals. These compensation signals are then applied to correct the detection signals from active pixels, creating a closed-loop feedback system that continuously adapts to noise conditions.
3Measurement precision
If binning circuits are used to process monitoring signals, then noise measurement accuracy is improved, but device complexity increases
Solution Approach 1:
Instead of processing every individual monitoring pixel signal separately, the binning circuit performs a partial action by combining multiple signals into an average. This approach provides sufficient noise measurement accuracy without the complexity of processing each signal individually, achieving the right level of precision for the application.
Data Source
AI summary
An image sensor compensates for noise. The image sensor includes a pixel array that includes a common monitor output line, a first monitoring pixel outputting a first monitoring signal, a second monitoring pixel outputting a second monitoring signal, and an active pixel configured to output a sensing signal based on an incident light. The image circuit also includes a binning circuit that receives the first and second monitoring signals through the common monitor output line and generates an average monitoring signal by performing binning on the first and second monitoring signals, and an analog-to-digital converter that detects an alternating current (AC) component of the average monitoring signal and couples the sampled AC component of the average monitoring signal to the sensing signal, thereby compensating for noise.