Image Sensor Sampling Circuitry for Sparse CFA Patterns

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional sampling and readout techniques are inefficient when applied to sparse Color Filter Array (CFA) patterns in electronic image sensors, which are used in digital cameras and other image capture devices.

Innovation Solution

The implementation of improved sampling and readout techniques specifically designed for image sensors with sparse CFA patterns, including a panchromatic checkerboard pattern, where pixels are configured to share common outputs and utilize column circuits to sample and read out pixels efficiently, allowing for binning of same-color and panchromatic pixels to enhance signal-to-noise ratio and frame rate.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional sampling and readout techniques are used with sparse CFA patterns, then pixel order is preserved in readout data, but sampling and readout efficiency deteriorates

Engineering Contradiction:
Improvepixel order preservationVSAvoidsampling and readout efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent segments the pixel array into distinct groups based on their CFA pattern characteristics (e.g., color pixels vs. panchromatic pixels). Each group is processed through dedicated column circuits that handle specific sampling and readout operations independently. This segmentation allows efficient processing of each pixel type while maintaining the ability to reconstruct the original pixel order through controlled readout sequencing.

Inventive Principle:
Principle #1Segmentation

2Quantity of substance

If all pixels in a row are sampled into column circuits simultaneously, then complete row data is captured, but overhead time increases

Engineering Contradiction:
Improvedata capture completenessVSAvoidoverhead time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent implements periodic action by dividing the sampling and readout process into multiple phases or cycles. Instead of sampling all pixels simultaneously and then reading them all, the system performs periodic sampling of pixel groups followed by staggered readout operations. This allows overlapping of sampling and readout operations across different pixel groups, reducing idle time and overhead while ensuring complete data capture.

Inventive Principle:
Principle #19Periodic action

3Illumination intensity

If sparse CFA patterns with panchromatic pixels are used, then light sensitivity is improved, but conventional sampling techniques become inefficient

Engineering Contradiction:
Improvelight sensitivityVSAvoidsampling technique efficiency
Core Design Contradiction:
Illumination intensityVSProductivity

Solution Approach 1:

The patent applies local quality by assigning different sampling and readout characteristics to different regions or types of pixels within the sparse CFA pattern. Panchromatic pixels, which have higher light sensitivity, are handled through dedicated column circuits with optimized sampling timing and readout sequences. Color pixels are processed through separate circuits with their own optimization parameters. This localized optimization ensures that each pixel type operates at peak efficiency while maintaining overall system performance.

Inventive Principle:
Principle #3Local quality

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

These techniques improve the efficiency of image capture by reducing overhead time and enhancing signal-to-noise ratio and frame rate, particularly suited for low-light conditions and fast autofocus applications.

Implementation Method 1

Each pixel comprises a photodiode that provides a charge packet proportional to the amount of incident light

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentEP2339829B1Sampling and readout of an image sensor
Publication Date: 2019.07.10 OMNIVISION TECHNOLOGIES INC
  • EP2339829B1 patent drawingFigure 1
  • EP2339829B1 patent drawingFigure 2
  • EP2339829B1 patent drawingFigure 3

AI summary

A CMOS image sensor or other type of image sensor comprises an array of pixels arranged in rows and columns, with the columns being separated into groups each comprising two or more columns that share a common output. The image sensor further comprises sampling and readout circuitry that includes, for each group of columns in the pixel array, a corresponding set of two or more column circuits. The sampling and readout circuitry is configured to sample the common output for each group of columns independently into one of the column circuits associated with that group, and to read out the common output for each group of columns as previously sampled into another of the column circuits associated with that group. The image sensor may be implemented in a digital camera or other type of image capture device.