Panchromatic Color Pixel Interpolation for Low-Light Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current digital imaging systems face limitations in low-light conditions due to the tradeoff between light sensitivity and color spatial resolution, as only a quarter of pixels in color filter arrays have maximum light sensitivity, compromising overall low-light performance and color fidelity.

Innovation Solution

A method is introduced to create a digital color image using a red-green-blue-panchromatic color filter array with an adaptive image processing strategy that interpolates panchromatic and color pixels, enhancing low-light sensitivity and color spatial resolution by interpolating high-resolution panchromatic and color difference images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If panchromatic pixels are used in color filter array, then light sensitivity is improved, but color spatial resolution deteriorates

Engineering Contradiction:
Improvelight sensitivityVSAvoidcolor spatial resolution
Core Design Contradiction:
Illumination intensityVSMeasurement precision

Solution Approach 1:

The image processing is segmented into multiple stages: first interpolating panchromatic pixels to create a high-resolution luminance image, then separately processing color difference components. This segmentation allows each component to be optimized independently, resolving the contradiction between light sensitivity (achieved through panchromatic interpolation) and color spatial resolution (maintained through separate color processing).

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Color difference images serve as an intermediary representation. Instead of directly interpolating color pixels which would lose spatial resolution, the patent computes color differences between panchromatic and color pixels, interpolates these differences, and then adds them back to the interpolated panchromatic image. This intermediary approach preserves both light sensitivity and color spatial resolution.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If more pixels are made panchromatic to improve low-light sensitivity, then overall low-light performance is improved, but color spatial resolution capability deteriorates

Engineering Contradiction:
Improvelow-light performanceVSAvoidcolor spatial resolution
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent changes the processing parameters dynamically based on local image characteristics. By computing gradient magnitudes and orientations in different directions, the interpolation adapts to edges and textures, maintaining color spatial resolution even when most pixels are panchromatic. This parameter adaptation resolves the contradiction between low-light performance and color resolution.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The interpolation process is made dynamic and adaptive rather than static. The algorithm evaluates local image structures and adjusts interpolation directions and weights accordingly, allowing the system to maintain high color spatial resolution despite having predominantly panchromatic pixels. This dynamic approach enables both improved low-light performance and preserved color resolution.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP1977613B1Interpolation of panchromatic and color pixels
Publication Date: 2015.10.07 OMNIVISION TECHNOLOGIES INC
  • EP1977613B1 patent drawingFigure 1
  • EP1977613B1 patent drawingFigure 2
  • EP1977613B1 patent drawingFigure 3

AI summary

A method for forming a final digital color image, includes: capturing an image using an image sensor having panchromatic pixels and color pixels corresponding to at least two color photoresponses; providing from the captured image a digital high resolution panchromatic image and a digital high resolution color differences image; and using the digital high resolution panchromatic image and the digital high resolution color differences image to produce the final digital high resolution full color image.