Image Sensor Pixel Merging for Low Light Sensitivity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image sensors have limitations in sensitivity and signal-to-noise ratio, particularly in low-light conditions, which affect the quality of captured images.

Innovation Solution

The image sensor design includes a light sensing array with adjacent light sensing pixels sharing source followers to collect and convert photoinduced charges into analog signals, and a control method to synthesize high dynamic range images by controlling clock, reset, and readout switches to optimize exposure and signal processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional image sensor design with individual readout circuits per pixel is used, then pixel-level signal processing capability is maintained, but sensitivity and signal-to-noise ratio deteriorate in low-light conditions

Engineering Contradiction:
Improvesignal-to-noise ratioVSAvoidreadout circuit structure
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple pixel signals (e.g., 2x2 or larger groups) into shared readout circuits called source followers. Instead of dedicating one readout circuit per pixel, multiple adjacent pixels share common source followers that collect and amplify their combined signals. This merging approach increases the effective signal level and improves signal-to-noise ratio by utilizing pooled photoinduced charges from multiple pixels, while reducing the overall number of readout circuits required.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The light sensing array is segmented into multiple light sensing units, where each unit contains a group of adjacent pixels that share common readout circuits. This segmentation allows the system to balance between pixel-level processing capability and circuit sharing benefits. Each light sensing unit can be independently controlled and read out, maintaining functional modularity while achieving improved sensitivity through circuit sharing within each unit.

Inventive Principle:
Principle #1Segmentation

2Productivity

If multiple readout circuits per pixel unit are implemented, then signal processing capability is improved, but circuit complexity and manufacturing difficulty increase

Engineering Contradiction:
Improvesignal processing capabilityVSAvoidcircuit integration
Core Design Contradiction:
ProductivityVSEase of manufacture

Solution Approach 1:

Multiple pixels within each light sensing unit share common source follower circuits, reducing the total number of readout circuits needed across the entire sensor array. This merging strategy maintains adequate signal processing capability by pooling signals from multiple pixels through shared circuits, while significantly reducing manufacturing complexity and improving ease of circuit integration compared to having dedicated circuits for every pixel.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If pixel signals are read out through individual vertical signal lines, then signal integrity is maintained, but sensitivity and signal-to-noise ratio are insufficient for low-light photography

Engineering Contradiction:
Improvesignal integrityVSAvoidsensitivity
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent merges signals from multiple pixels before they reach the output, using shared source follower circuits to collect and amplify combined photoinduced charges. This merging occurs while maintaining signal integrity through controlled charge collection and sequential readout mechanisms. The combined signal from multiple pixels achieves higher sensitivity and better signal-to-noise ratio for low-light conditions, while the systematic readout approach preserves signal fidelity.

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 design enhances sensitivity and signal-to-noise ratio, allowing for improved image quality and the ability to capture high dynamic range images, especially in low-light environments.

Implementation Method 1

light sensing pixels...receive the same color of light...collects photoinduced charges of the plurality of light sensing pixels

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentEP3313069B1Image sensor, control method, and electronic device
Publication Date: 2020.12.16 GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD
  • EP3313069B1 patent drawingFigure 1~2
  • EP3313069B1 patent drawingFigure 3
  • EP3313069B1 patent drawingFigure 4

AI summary

Disclosed is an image sensor. The image sensor comprises a photosensitive unit array, a filter unit array, and multiple converting units. The filter unit array is arranged on the photosensitive unit array. Each filter unit covers corresponding one photosensitive unit. The photosensitive units comprise multiple photosensitive pixels. The converting units comprise at least two source followers. At least one source follower is connected to the multiple photosensitive pixels. Also disclosed are a control method and an electronic device. The image sensor of embodiments of the present invention, where a same filter unit corresponds to the multiple photosensitive pixels and at least one source follower is connected to the multiple photosensitive pixels, compared with single photosensitive pixel output, is capable of acquiring an increased amount of photo-generated charge, thus not only increasing sensitivity and signal-to-noise ratio by means of hardware improvement, but also allowing composition of high dynamic range images by means of controlling output and reading of photosensitive pixel exposure.