Bayer Mask Image Decoding with Slope-Adaptive Interpolation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing Bayer mask decoders suffer from cross-color and 'barber's pole' artefacts, particularly on the sparsely sampled red and blue components, despite advancements in non-adaptive linear filters and slope adaptive filters.

Innovation Solution

A method and apparatus that utilize a two-dimensional filter to generate a full-resolution grid of green samples, followed by color difference signal processing, slope and activity calculation, and directional filtering to produce decoded red and blue signals, incorporating a slope index for steered interpolation and a slope consistency measure to reduce artefacts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If basic Bayer mask decoders are used, then decoding simplicity is maintained, but cross-color and barber's pole artefacts occur

Engineering Contradiction:
Improvedecoder complexityVSAvoidcross-color and barber's pole artefacts
Core Design Contradiction:
Device complexityVSObject-generated harmful factors

Solution Approach 1:

The patent implements adaptive filtering where filter parameters dynamically adjust based on local image characteristics. The slope measure and activity measure allow the decoder to adapt its behavior to different regions, switching between different filtering strategies to minimize artifacts while maintaining decoding effectiveness.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes filtering parameters based on local image properties. By computing slope measures and activity measures, the system adjusts filter strength and type according to local gradient characteristics, thereby reducing artifacts in different regions without requiring a completely complex decoder architecture.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If non-adaptive linear filters are used for decoding, then green component decoding performance is improved, but cross-color artefacts increase on red and blue components

Engineering Contradiction:
Improvegreen component decoding accuracyVSAvoidcross-color artefacts on red and blue
Core Design Contradiction:
Measurement precisionVSObject-generated harmful factors

Solution Approach 1:

The patent applies different filtering strategies to different color components based on their sampling characteristics. Green components receive non-adaptive linear filtering to maintain decoding accuracy, while red and blue components receive adaptive filtering based on slope and activity measures to prevent cross-color artifacts.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent dynamically adjusts filtering parameters for red and blue components based on local image characteristics computed from the green component. By using slope measures and activity measures, the system adapts the filtering strength to prevent cross-color artifacts while preserving the benefits of linear filtering for the green component.

Inventive Principle:
Principle #35Parameter changes

3Object-generated harmful factors

If slope adaptive filters are used, then artefact reduction is achieved, but computational complexity and processing time increase

Engineering Contradiction:
Improveartefact reductionVSAvoidprocessing time
Core Design Contradiction:
Object-generated harmful factorsVSLoss of time

Solution Approach 1:

The patent computes slope measures and activity measures selectively rather than for all pixels. By applying adaptive filtering only where needed based on local gradient characteristics, the system reduces artifacts without requiring full computational processing across the entire image, thereby reducing processing time.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3547249B1Decoding a bayer-mask or like coded image
Publication Date: 2021.08.25 GRASS VALLEY LTD
  • EP3547249B1 patent drawingFigure 1
  • EP3547249B1 patent drawingFigure 2
  • EP3547249B1 patent drawingFigure 3~5

AI summary

A Bayer-mask image is decoded by forming a decoded green array; calculating a slope at each pixel of the array, expressed as an angle; calculating an activity at each pixel; converting the slope angle and the activity into a complex number for each pixel, of modulus equal to the activity and argument equal to twice the slope angle; expressing said complex numbers in Cartesian coordinates to form a Cartesian slope signal and filtering the Cartesian slope signal with a linear spatial filter to derive a slope measure. Blue - green and red-green values are calculated and interpolated using a slope-adaptive interpolation filter steered by said slope measure.