DPCM Codec Bounded Reconstruction Error for Flat-Panel Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression technologies often result in undesirable reconstruction errors, particularly motion blur and ghosting effects in flat-panel displays, due to the inability to ensure non-negative and bounded reconstruction errors, which affects the performance of overdrive systems used for image compensation.
Innovation Solution
The proposed system employs a method of image compression and storage that maps original samples to mapped samples based on bit depth and maximum allowed error, applying modulo addition and quantization to ensure non-negative and bounded reconstruction errors, particularly for high-priority gray values, by using appropriate quantization band sizes and added bias.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If conventional image compression is used, then storage efficiency is improved, but reconstruction error becomes unbounded and negative values occur causing motion blur and ghosting effects
Solution Approach 1:
The patent changes the parameter space by applying a mapping function that transforms original pixel values to a new domain where compression is performed. This parameter transformation allows the compression algorithm to operate in a space where reconstruction errors are inherently non-negative and bounded, resolving the contradiction between compression efficiency and reconstruction reliability
Solution Approach 2:
The patent introduces an intermediary mapping function as a mediator between the original image data and the compression process. This mapping layer transforms the data into a form that can be compressed while guaranteeing that the reconstruction errors remain within acceptable bounds and non-negative, thus enabling both efficient storage and reliable reconstruction
2Measurement precision
If quantization band size is reduced to improve reconstruction quality, then encoding complexity and processing time increase
Solution Approach 1:
By changing the parameter space through the mapping function, the patent enables the use of larger quantization bands in the transformed domain while maintaining acceptable reconstruction quality. This parameter transformation effectively decouples the relationship between quantization band size and reconstruction quality, allowing faster encoding without significant quality loss
3Loss of substance
If DPCM compression is applied to reduce data size, then reconstruction error for certain gray levels increases causing display artifacts
Solution Approach 1:
The patent applies parameter changes by transforming the data into a mapped space before applying DPCM compression. In this transformed space, the differential values have different statistical properties that make them more suitable for DPCM, allowing effective data size reduction while maintaining or improving gray level reconstruction accuracy for important display values
Solution Approach 2:
The patent implements local quality by ensuring that the mapping and compression process specifically preserves the accuracy of important gray levels that are critical for display quality. The system prioritizes the reconstruction accuracy of certain gray level ranges over others, applying different compression aggressiveness to different parts of the data based on their importance
Data Source
AI summary
A method of compressing a frame in an image compression and storage system includes mapping an original sample to a mapped sample based on a bit depth of the original sample and a maximum allowed error, determining a residue of the mapped sample based on a mapped previous reconstructed sample, applying a modulo addition to the residue to generate a biased residue, quantizing the biased residue based on the maximum allowed error to generate a quantized biased residue, and encoding a value corresponding to the quantized biased residue to generate an encoded value.