LDR to HDR Image Enhancement via Tone Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing images and videos often have lower dynamic ranges than new high-performance display devices can handle, limiting their ability to accurately reproduce real scenes without noticeable artefacts.
Innovation Solution
Methods are developed to enhance the dynamic range of lower-dynamic-range (LDR) image data by linearizing pixel values, stretching contrast, applying a brightness enhancement function, and using edge-stopping components to avoid artefacts, allowing for real-time processing of video frames to produce higher-dynamic-range (HDR) output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If image data is processed to enhance dynamic range, then the dynamic range of the output image is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent applies tone mapping operations and brightness enhancement functions in advance to convert LDR image data into HDR-compatible format. By pre-processing the image data through these transformation functions, the system prepares the content for HDR display without requiring real-time processing during playback, thus reducing processing time while maintaining enhanced dynamic range.
2Illumination intensity
If brightness enhancement is applied to LDR image data, then the dynamic range is improved, but image artefacts may be introduced
Solution Approach 1:
The patent transforms image data by changing the luminance parameter representation from LDR to HDR space using tone mapping functions. This parameter transformation allows the image to accommodate the wider dynamic range of HDR displays while using mathematically controlled mapping functions that preserve image quality and prevent artefact generation during the conversion process.
Solution Approach 2:
The patent introduces tone mapping functions as intermediary mathematical transformations between LDR input and HDR output. These intermediary functions serve as a bridge that smoothly maps the limited LDR luminance values into the extended HDR luminance space, preventing direct abrupt transformations that would cause image artefacts while achieving the desired dynamic range enhancement.
3Illumination intensity
If contrast stretching is applied to enhance dynamic range, then the brightness range is improved, but noise and artefacts increase
Solution Approach 1:
The patent applies contrast stretching through gamma correction and tone mapping parameter transformations that map LDR luminance values to HDR luminance space. By using controlled mathematical parameter changes rather than simple linear scaling, the system expands the brightness range while maintaining image quality and preventing noise amplification that would occur with aggressive contrast stretching.
Data Source
AI summary
Methods and apparatus according to various aspects take as input image data in a lower-dynamic-range (LDR) format and produce as output enhanced image data having a dynamic range greater than that of the input image data (i.e. higher-dynamic range (HDR) image data). In some embodiments, the methods are applied to video data and are performed in real-time (i.e. processing of video frames to enhance the dynamic range of the video frames is completed at least on average at the frame rate of the video signal).