Image Processing Gain Data for HDR-LDR Compatibility

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current devices with low dynamic range (LDR) 8-bit displays cannot fully utilize the extra precision of high dynamic range (HDR) image content, as the differences between HDR and LDR images are not readily visible, limiting the enhancement of image quality for both LDR and HDR displays.

Innovation Solution

A method and computing device that generate gain-related data based on both HDR and LDR versions of an image, allowing users to control the appearance of the output image through a gain-related control parameter, enabling flexible display on both HDR and LDR devices by applying effective gain-related data to the image data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If HDR image data with higher bit precision is displayed on LDR 8-bit displays, then the image quality enhancement is limited, but the HDR content precision is wasted

Engineering Contradiction:
Improveimage precisionVSAvoiddisplay compatibility
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The image data is segmented into multiple versions with different dynamic ranges and bit precisions. The system divides the HDR image into a high dynamic range version (first version) and a standard dynamic range version (second version), allowing each version to be optimized for its intended display type while maintaining both in the same image file

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically adapts the image display characteristics based on the display device capabilities. By providing multiple versions and using gain-related data with control parameters, the image can be dynamically adjusted to suit either LDR or HDR displays, making the system flexible and adaptive rather than static

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If JPEG XT format with higher bit precision is used, then HDR image quality is improved, but compatibility with current LDR displays is reduced

Engineering Contradiction:
Improveimage precisionVSAvoiddisplay compatibility
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The image file is designed to serve multiple functions and multiple display types simultaneously. By incorporating both HDR and SDR versions along with gain-related data, a single image file can be universally displayed on both LDR and HDR devices, eliminating the need for separate image files for different display types

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

Solution Approach 2:

The system performs preliminary processing to create multiple versions of the image and embed gain-related data before display. The gain-related data and control parameters are pre-calculated and stored in the image file, allowing the display device to automatically select and apply the appropriate processing without requiring real-time conversion

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If gain-related data processing is applied to enhance LDR display quality, then image quality on LDR displays is improved, but processing complexity increases

Engineering Contradiction:
Improveimage qualityVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Gain-related data serves as an intermediary between the HDR image content and the LDR display device. This intermediary contains pre-calculated adjustment information that mediates the transformation from HDR to SDR, simplifying the processing required at the display device while still achieving quality enhancement

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10783620B2Image processing
Publication Date: 2020.09.22 ARM LTD
  • US10783620B2 patent drawing
  • US10783620B2 patent drawing
  • US10783620B2 patent drawing

AI summary

A method is provided. The method includes receiving image data representing an image. The image data includes first image data representing a first version of the image. The image data includes second image data representing a second version of the image. The method includes generating gain-related data based on the first image data and the second image data. The method includes generating effective gain-related data at least based on the gain-related data and a gain-related control parameter. The method includes applying the effective gain-related data to the image data to generate output image data representing an output version of the image. A computing device is also provided.