Gamut Mapping Unit for Image Sensor Hue Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for compressing image sensor pixel component values to match display gamut, such as clipping and scaling, result in poor hue accuracy and perceptually detrimental artefacts, as they either alter colors within the display gamut or fail to convert negative values effectively.

Innovation Solution

A gamut mapping method that adjusts negative input component values to non-negative values and compensates other values to preserve hue and saturation, allowing conversion from an extended RGB format to an unextended format like sRGB, using a gamut mapping unit with adjustment and scaling blocks to ensure output values fall within the valid range.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If clipping is applied to component values outside display gamut, then component values are brought into display gamut range, but hue accuracy deteriorates and colour flatness occurs

Engineering Contradiction:
Improvecomponent value range complianceVSAvoidhue accuracy
Core Design Contradiction:
Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The patent transforms the problem by changing the parameter space - converting from extended RGB format (with negative values) to unextended RGB format (with non-negative values only). This parameter transformation allows the system to handle out-of-gamut colors differently, preserving hue relationships while ensuring compliance with display gamut constraints through format conversion rather than simple clipping

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments the gamut mapping process into distinct stages: first converting from extended to unextended format, then applying gamut compression. This segmentation allows each stage to address specific aspects of the problem independently, preventing the hue accuracy deterioration that occurs when clipping is applied directly to the original component values

Inventive Principle:
Principle #1Segmentation

2Manufacturing precision

If scaling is applied to reduce component values, then all component values are reduced by the same factor, but intensity of colours within display gamut is reduced

Engineering Contradiction:
Improvecomponent value range complianceVSAvoidcolour intensity
Core Design Contradiction:
Manufacturing precisionVSIllumination intensity

Solution Approach 1:

The patent changes the parameter representation by converting to unextended RGB format where negative values become positive. This transformation fundamentally alters how scaling operates - instead of uniformly reducing all values and losing intensity information, the system can apply more selective adjustments that preserve the intensity of colors already within the display gamut while handling out-of-gamut colors appropriately

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If extended RGB format with negative values is used, then complete colour representation is achieved, but compatibility with display formats deteriorates

Engineering Contradiction:
Improvecolour representation rangeVSAvoiddisplay format compatibility
Core Design Contradiction:
Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent applies preliminary action by performing the format conversion from extended to unextended RGB early in the processing pipeline, before final display output. This preliminary transformation ensures that all subsequent processing works with display-compatible formats while the original extended format information is preserved during intermediate stages, maintaining both versatility and compatibility

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS9444975B2Image sensor gamut mapping
Publication Date: 2016.09.13 IMAGINATION TECH LTD
  • US9444975B2 patent drawing
  • US9444975B2 patent drawing
  • US9444975B2 patent drawing

AI summary

Gamut mapping is applied to a pixel to thereby convert input component values (e.g. RGB values) of the pixel to corresponding output component values of the pixel. The input component values have a first valid range which includes some negative values and the output component values have a second valid range which does not include negative values. The gamut mapping unit determines whether a first input component value of the pixel is negative, and if so: (i) applies an adjustment to that component value such that it is not negative; and (ii) adjusts one or more other input component value(s) of the pixel based on the adjustment applied to the first input component value to thereby compensate for the adjustment applied to the first input component value. The output component values of the pixel are based on the adjusted input component values of the pixel.