Heterogeneous Camera Array Imaging for Parallax Depth and Low-Light SNR

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image sensors in mobile devices face performance constraints such as low light sensitivity, limited dynamic range, and reduced signal-to-noise ratio (SNR) due to small pixel size and limited light gathering capability.

Innovation Solution

A distributed approach using a camera array with multiple heterogeneous imagers, each with custom filters, optics, and operating parameters, captures images that are then processed to generate an enhanced image with improved resolution and performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Volume of moving object

If pixel size is reduced to fit mobile device constraints, then device compactness is improved, but light gathering capability and SNR deteriorate

Engineering Contradiction:
Improvedevice sizeVSAvoidsignal to noise ratio
Core Design Contradiction:
Volume of moving objectVSMeasurement precision

Solution Approach 1:

The patent divides the image sensor into multiple independent photodetectors (e.g., 2x2 array = 4 photodetectors) within a compact footprint. Each photodetector can be independently controlled with different integration times, allowing some to accumulate more light for better SNR while others capture faster-changing information, resolving the contradiction between small size and signal quality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic control of integration time for each photodetector independently. During a single exposure period, different photodetectors can have different integration durations (e.g., 10ms, 20ms, 30ms), enabling adaptive light gathering optimization without increasing device size or pixel count.

Inventive Principle:
Principle #15Dynamics

2Illumination intensity

If integration time is increased to improve light sensitivity, then low light performance is improved, but dynamic range and temporal resolution deteriorate

Engineering Contradiction:
Improvelow light sensitivityVSAvoidtemporal resolution
Core Design Contradiction:
Illumination intensityVSSpeed

Solution Approach 1:

The patent dynamically varies integration time across multiple photodetectors within a single exposure period. Some photodetectors use long integration times (e.g., 30ms) for low-light sensitivity, while others use short integration times (e.g., 10ms) for temporal resolution, with all readings combined to produce a single image that benefits from both approaches.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent adds a temporal dimension to the imaging process by allowing different integration times within a single exposure period. This transforms the traditional single integration-time model into a multi-temporal-dimensional model, enabling simultaneous optimization of both low-light performance and temporal resolution through computational combination of readings.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Loss of information

If color filters are added to achieve color imaging, then color information is improved, but light gathering capability and SNR deteriorate

Engineering Contradiction:
Improvecolor informationVSAvoidsignal to noise ratio
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The patent segments the color filtering function across multiple photodetectors rather than placing filters on every pixel. Each photodetector can be assigned a specific color filter (e.g., red, green, blue, or panchromatic), and the system uses computational methods to reconstruct full-color images from the segmented readings, improving light gathering while maintaining color information.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent makes photodetectors multi-functional by allowing them to serve different color channels at different times or in different spatial positions. Panchromatic photodetectors can capture all wavelengths for luminance information, while color-filtered photodetectors provide chroma information, with the system adaptively using each detector's strengths to overcome the SNR penalty of color filtering.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Measurement precision

If multiple camera arrays are used to improve image quality, then dynamic range and resolution are improved, but device complexity and manufacturing cost deteriorate

Engineering Contradiction:
Improveimage qualityVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple photodetector readings taken during a single exposure period into one composite image. By combining information from photodetectors with different integration times and spectral responses captured simultaneously, the system achieves high dynamic range and resolution without requiring multiple separate camera modules, thereby reducing device complexity and manufacturing cost.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary capture of multiple temporal and spectral samples within a single exposure period, then processes these pre-captured readings computationally to generate the final high-quality image. This preliminary multi-sample capture approach eliminates the need for multiple physical camera arrays, simplifying the system while maintaining image quality.

Inventive Principle:
Principle #10Preliminary action

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

The camera array achieves enhanced image quality with increased dynamic range, improved low light sensitivity, and higher SNR, effectively addressing the performance limitations of conventional image sensors in mobile devices.

Implementation Method 1

The image sensor consists of pixels that generate signals upon receiving light via the optical element

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentEP4336447B1Capturing and processing of images using monolithic camera array with heterogeneous imagers
Publication Date: 2025.05.07 ADEIA IMAGING LLC
  • EP4336447B1 patent drawingFigure 1~2A
  • EP4336447B1 patent drawingFigure 2B~3B
  • EP4336447B1 patent drawingFigure 3C~4

AI summary

A method for estimating distance to an object in a scene using an imaging system (400) comprising a camera array (410) with multiple imagers (540), an image processing pipeline module (420) and a controller (440). The image processing pipeline module (420) comprises an upstream pipeline processing module (510), an image pixel correlation module (514), and a parallax confirmation and measurement module (518). Images (412), including a current image, captured by the multiple imagers (540) are provided to the upstream pipeline processing module (510) for processing. The image pixel correlation module (514) aligns portions of images, including the current image, captured by different imagers of the multiple imagers to compensate for parallax. The parallax confirmation and measurement module (518) processes the current image to detect and measure the parallax. Distance to an object in the scene is determined using the measured parallax.