Image Sensor Pixel Binning with Shared Readout Circuits

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Fast image sensors face challenges in achieving high capture rates through pixel grouping without oversizing read circuit components, which affects storage element charging and operational efficiency.

Innovation Solution

The image sensor design includes a matrix of pixels with photodetectors and read circuits that can form macro-pixels by coupling the inputs of read circuits in parallel and connecting their outputs to a single storage element, utilizing active amplification elements to enhance bandwidth and rapid charging of storage capacitors without component oversizing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If pixel grouping (binning) is implemented to increase capture rate, then productivity is improved, but the read circuit components must be oversized which increases device complexity and affects storage element charging

Engineering Contradiction:
Improvecapture rateVSAvoidread circuit component size
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements dynamic reconfiguration of the pixel array where pixels can be dynamically grouped into macro-pixels during high-speed capture modes. Switching elements enable runtime reconfiguration of read circuit connections, allowing the system to adapt between individual pixel operation and macro-pixel operation without permanent structural changes, thus avoiding oversizing while enabling high productivity when needed

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent segments the read circuit functionality by creating shared read circuits that can serve multiple pixels when grouped. Instead of each pixel having a dedicated oversized read circuit, the system segments the reading function across shared circuits that handle macro-pixel data, reducing individual component size requirements while maintaining high capture rate capability

Inventive Principle:
Principle #1Segmentation

2Productivity

If read circuit components are oversized to achieve high capture rates during pixel grouping, then productivity is improved, but manufacturing precision and operational efficiency deteriorate

Engineering Contradiction:
Improvecapture rateVSAvoidcomponent sizing accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent changes operational parameters by switching between different read circuit configurations based on capture mode. During high-speed burst mode, pixels are grouped and shared read circuits operate with adjusted timing and integration parameters. During normal mode, individual pixels use their dedicated circuits with standard parameters, maintaining manufacturing precision without requiring oversized components for all operating conditions

Inventive Principle:
Principle #35Parameter changes

3Productivity

If pixels are grouped into macro-pixels to increase capture rate, then productivity is improved, but storage element charging time becomes insufficient

Engineering Contradiction:
Improvecapture rateVSAvoidstorage element charging time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent merges multiple photodetector signals at the input stage of shared read circuits before amplification and integration. By combining signals early in the processing chain and using capacitive integration within the shared CTIA, the system accumulates charge from multiple pixels simultaneously, providing sufficient charging time even during high-speed macro-pixel operation without compromising capture rate

Inventive Principle:
Principle #5Merging (Combining)

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 configuration allows for high capture rates during pixel grouping while maintaining efficient storage element charging and operational efficiency, even when not in grouping mode, without the need for oversized read circuit components.

Implementation Method 1

Each pixel comprises a photodetector and a readout circuit associated with the photodetector. The photodetector corresponds for example to a photodiode.

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Implementation Method 2

The read circuit is connected to a memorization element which can correspond to a storage capacitor when the signal delivered by the read circuit is then processed in an analog manner.

Methodology Applied
Scientific EffectCapacitance: Capacitance

Data Source

PatentEP3629573B1Quick image sensor with pixel binning
Publication Date: 2021.04.21 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES
  • EP3629573B1 patent drawingFigure 1~2
  • EP3629573B1 patent drawingFigure 3~4
  • EP3629573B1 patent drawingFigure 5

AI summary

An image sensor comprising a pixel array (110) each pixel comprising: - a photodetector (112); - a readout circuit (160) coupled to the photodetector (112) and comprising an active amplifier element (114); and: - the sensor comprises means (122, 124, 126, 128, 130) for grouping pixels configured to form macro-pixels such that in each macro-pixel, the inputs of the readout circuits of at least two pixels are coupled to each other, and the outputs of the readout circuits of said pixels are coupled to the input of a single storage element comprising a capacitor (119); - the output of each readout circuit is coupled to an input of a storage element different from those of the other pixels; - the grouping means include switches (130) for connecting the reading circuits (160) interposed between the reading circuits and the storage elements.