Partial Code Allocation Functions for HDR Image Encoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current technologies face challenges in efficiently encoding and processing High Dynamic Range (HDR) images due to suboptimal code allocation functions, leading to image degradation and quality issues when trying to fit HDR data into existing 8-bit or 10-bit formats, especially in color grading and display applications.

Innovation Solution

The use of partial code allocation functions that can be sequentially applied to map linear luminance values to luma codes, allowing for a flexible and efficient representation of HDR images using fewer bits, while maintaining image quality and enabling compatibility with existing LDR systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If conventional code allocation functions are used to map HDR linear luminance values to luma codes, then the HDR image data can be encoded into reduced bit formats (8-bit or 10-bit), but image degradation and quality issues occur due to suboptimal code allocation

Engineering Contradiction:
Improvebit depthVSAvoidimage quality
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The code allocation function is divided into multiple sequential partial functions (first partial function, second partial function, etc.). Each partial function processes a specific segment of the luminance range, allowing optimized code distribution across different luminance regions. This segmentation enables the system to maintain image quality while reducing bit depth by applying different mapping strategies to different parts of the luminance spectrum.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the parameters of the code allocation function by using multiple partial functions with different characteristics. Each partial function can have different mapping properties (e.g., different gamma values, different luminance range coverage) to optimize the distribution of code values across the reduced bit depth range, thereby maintaining image quality during compression.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If existing LDR systems and interfaces are reused for HDR image processing, then compatibility and ease of operation are improved, but the systems face challenges in efficiently handling the increased dynamic range

Engineering Contradiction:
Improvecompatibility with LDR systemsVSAvoidefficiency in processing HDR data
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent makes existing LDR systems universal by designing a code allocation approach that can handle both LDR and HDR images. The sequential partial functions are configured to map HDR luminance values into the standard LDR code range, allowing the same system infrastructure to process both dynamic range types without requiring separate dedicated HDR processing paths, thus achieving compatibility while maintaining efficiency.

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

Solution Approach 2:

The sequential partial functions act as intermediaries between HDR linear luminance values and the final luma codes suitable for LDR systems. These partial functions transform the HDR data through intermediate representation stages, enabling existing LDR interfaces to handle HDR content efficiently without direct modification to the underlying system architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If a single code allocation function is used for HDR mapping, then the device complexity is reduced, but the image quality and adaptability to different luminance ranges deteriorate

Engineering Contradiction:
Improvenumber of mapping functionsVSAvoidcode allocation accuracy
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

Instead of using a single complex code allocation function, the patent segments the mapping process into multiple simpler partial functions. Each partial function handles a specific aspect of the luminance-to-code mapping, and they are applied sequentially. This segmentation maintains manageable device complexity while improving code allocation accuracy through specialized processing at each stage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent combines multiple partial functions to create a comprehensive code allocation solution. By merging the results of sequential partial function applications, the system achieves high code allocation accuracy that would be difficult to obtain with a single function, while keeping each individual partial function relatively simple and manageable.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3022902B1Method and apparatus to create an EOTF function for a universal code mapping for an HDR image
Publication Date: 2019.11.20 KONINKLIJKE PHILIPS NV
  • EP3022902B1 patent drawingFigure 1~3
  • EP3022902B1 patent drawingFigure 4~5
  • EP3022902B1 patent drawingFigure 6~7

AI summary

To enable better encoding of the currently starting to appear high dynamic range images for use in full high dynamic range technical systems (containing an HDR display, and e.g. in an HDR grading application), we invented a method of constructing a code allocation function for allocating pixel colors having pixel luminances to codes encoding such pixel luminances, in which the step of determining the code allocation function to be applied to at least one color coordinate of the pixel to obtain a code value, comprises constructing that function out of at least two partial functions, and similar methods at a receiving side, and apparatuses, and signals for communicating between the two sites.