Readout Circuit Calibration for Column Fixed Pattern Noise

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional techniques fail to effectively address column fixed pattern noise (CFPN) in image capturing devices, particularly at low light levels and varying gain conditions, leading to image degradation.

Innovation Solution

A method involving calibration of readout circuits connected to programmable gain amplifiers and analog-to-digital converters, where mismatches are identified and corrected by disconnecting the readout circuit from the pixel output, measuring outputs at different switch configurations, and storing calibration information to adjust signal processing accordingly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional calibration techniques are used that account for combined CFPN and pixel imprecision, then general image degradation is addressed, but CFPN-induced degradation at low light levels is not effectively corrected

Engineering Contradiction:

Improveimage quality at low light levelsVSAvoidability to handle different lighting conditions

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent segments the calibration process into two distinct components: pixel-level calibration and readout circuit-level calibration. By separating these calibration functions and performing them at different stages, the system can specifically address CFPN in readout circuits independently from pixel imprecision, thereby improving low light level image quality without being constrained by conventional combined calibration approaches

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs readout circuit calibration during a preliminary calibration mode before normal operation. By calibrating the readout circuits in advance when no incident light is present, the system establishes baseline correction data that specifically targets CFPN. This preliminary calibration enables the system to effectively correct CFPN-induced degradation during subsequent low light level imaging operations

Inventive Principle:

Principle #10Preliminary action

2Illumination intensity

If gain amplifiers are used to boost sensor signals for higher contrast, then image contrast is improved, but the impact of CFPN is also increased

Engineering Contradiction:

Improveimage contrastVSAvoidCFPN impact

Core Design Contradiction:

Illumination intensityVSObject-affected harmful factors

Solution Approach 1:

The patent implements a feedback mechanism where calibration data obtained during calibration mode is stored and then applied during normal operation to correct readout circuit outputs. This feedback loop allows the system to compensate for CFPN that is amplified along with the signal, thereby maintaining image contrast improvement while reducing the visible impact of CFPN through active correction based on stored calibration characteristics

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent performs preliminary calibration of readout circuits at different gain settings during calibration mode. By pre-characterizing the CFPN behavior at each gain level and storing this calibration data, the system can apply appropriate correction factors during normal operation. This preliminary characterization enables the system to maintain high contrast images while compensating for gain-dependent CFPN amplification

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If conventional calibration methods are used that do not separate readout circuit calibration from pixel calibration, then device complexity is reduced, but measurement precision for CFPN correction is insufficient

Engineering Contradiction:

ImproveCFPN correction accuracyVSAvoidcalibration process complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the calibration process into distinct phases: a calibration mode for characterizing readout circuit CFPN and a normal operation mode for applying corrections. By separating these functions temporally and functionally, the system achieves high CFPN correction precision without requiring permanently complex hardware, as the additional calibration infrastructure is only active during the calibration phase

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements periodic calibration cycles where the system alternates between calibration mode and normal operation mode. During periodic calibration intervals, the readout circuits are calibrated to update CFPN correction data. This periodic approach maintains high correction accuracy over time while avoiding the need for continuous complex calibration operations, thereby balancing precision with operational simplicity

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS8830361B2Method of reducing column fixed pattern noise

Publication Date: 2014.09.09 TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD

AI summary

A method of reducing column fixed pattern noise including calibrating a readout circuit, wherein the readout circuit is electrically connected to at least one programmable gain amplifier and an analog-to-digital converter. Calibrating the readout circuit includes electrically disconnecting the readout circuit from a pixel output and electrically connecting a pixel reset input of the readout circuit to a pixel output signal input of the readout circuit. Calibrating the readout circuit further includes comparing a measured output of the readout circuit to a predetermined value and storing the comparison result in a non-transitory computer readable medium. The method further includes operating the readout circuit, the operating the readout circuit includes receiving a pixel sample signal and outputting a calibrated output based on an operating output and the stored comparison result.