SYUV Image Compression via Luma Edge Preservation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image data compression techniques in YUV formats face inefficiencies due to sub-sampling methods that do not effectively utilize chroma and luma values, leading to perceptible losses in image quality.
Innovation Solution
The proposed solution involves sub-sampling luma and chroma values using specific patterns (Bayer and 4:2:2/4:2:0) and compressing chroma values using DPCM, while incorporating edge information for discarded luma values, resulting in an efficient SYUV format that maintains perceptual losslessness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If chroma sub-sampling is implemented with traditional YUV formats, then data compression is achieved, but image quality suffers perceptible losses
Solution Approach 1:
The patent segments the image data into luma (Y) and chroma (UV) components, applying different sub-sampling strategies to each. The luma values are sub-sampled at a finer rate than traditional chroma sub-sampling, while chroma values are sub-sampled more aggressively. This segmentation allows the system to prioritize luminance information for image quality while reducing chrominance data for compression, resolving the contradiction between compression ratio and image quality.
Solution Approach 2:
The patent applies different quality levels to different components of the image data. Specifically, luma values retain higher local quality with less aggressive sub-sampling, while chroma values accept lower local quality with more aggressive sub-sampling. This local quality differentiation maintains perceptual image quality where it matters most (luma) while achieving compression through chroma reduction.
2Productivity
If luma values are sub-sampled aggressively, then compression efficiency increases, but edge information is lost
Solution Approach 1:
The patent performs preliminary identification of edge information in the luma data before sub-sampling. By detecting edges in advance, the system can preserve these critical features during subsequent sub-sampling operations, preventing loss of important structural information while still achieving compression efficiency through aggressive luma sub-sampling.
Solution Approach 2:
The patent incorporates feedback mechanisms that analyze the sub-sampled luma data to identify and recover edge information that may have been lost during compression. This feedback loop allows the system to adjust reconstruction parameters to preserve edge integrity, ensuring that compression efficiency gains do not come at the cost of critical structural information.
Data Source
AI summary
Chroma values in image data may be sub-sampled, thereby obtaining sub-sampled chroma values. The sub-sampled chroma values may be compressed, thereby obtaining compressed, sub-sampled chroma values. Luma values in the image data may be compressed, thereby obtaining sub-sampled luma values. Edge information for the luma values that are discarded as part of the luma sub-sampling operation may be determined.