Full Green Interpolation with Downsampling for Bayer Remosaicing

Solution Overview

Problem

Existing image sensors face challenges in producing high-quality remosaiced Bayer images with minimal artifacts and blurring, particularly in full green interpolation due to the reduction in pixel size and microlens manufacturing difficulties.

Innovation Solution

A method for full green interpolation involving down sampling and up sampling processes to create a full interpolated green image, using linear combinations of surrounding green pixel values to reduce artifacts and improve image quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If pixel size is reduced to increase image sensor resolution, then image sensor can capture more pixels (e.g., 200 MP), but microlens manufacturing becomes relatively harder and light sensitivity decreases

Engineering Contradiction:
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent merges multiple pixels (2x2 or 4x4 pixels) under a single microlens, forming a pixel group. This allows the microlens to maintain a reasonable size and manufacturing feasibility while effectively increasing the light collection area. The microlens covers multiple pixels simultaneously, improving light sensitivity without requiring proportionally smaller microlenses for each individual pixel.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent transitions from covering one pixel per microlens to covering multiple pixels (2x2 or 4x4 arrangement) under a single microlens. This dimensional change in the coverage area allows the system to maintain microlens manufacturability while effectively increasing the light-gathering capability and resolving the contradiction between pixel density and microlens manufacturing difficulty.

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

2Measurement precision

If full green interpolation is performed in remosaicing Bayer pattern, then image quality improves with less artifact and less blurred image, but processing complexity increases

Engineering Contradiction:
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs down-sampling of the Bayer pattern image before conducting green channel interpolation. By reducing the resolution first, the subsequent interpolation operation becomes computationally less intensive. After interpolation, the image is up-sampled back to the original resolution. This preliminary down-sampling action reduces the overall processing complexity while maintaining the quality benefits of full green interpolation.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If 4x4 pixel minimum repeating unit is used for quad phase detection, then phase difference detection capability is improved, but remosaicing complexity increases

Engineering Contradiction:
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the 4x4 pixel minimum repeating unit into four separate 2x2 Bayer pattern blocks. Each block can be independently processed during remosaicing, which simplifies the overall remosaicing operation compared to treating the entire 4x4 unit as a single entity. This segmentation maintains the phase detection capability while reducing remosaicing complexity through modular processing.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS12389132B2Full green interpolation in remosaicing Bayer pattern
Publication Date: 2025.08.12
  • US12389132B2 patent drawing
  • US12389132B2 patent drawing
  • US12389132B2 patent drawing

AI summary

A method for full interpolating green pixels from an input image having a first minimum repeating unit comprising 4×4 pixels, where 2×2 pixels of same color are grouped together, comprises down sampling of the input image to a first down sampled image, down sampling of the input image to a second down sampled image, and interpolating green pixels resulting in an interpolated green down sampled image, where the interpolating uses jointly the first down sampled image and the second down sampled image. The interpolated green pixels in an interpolated green down sampled image are further up sampled resulting in a full interpolated green image.