Decoder Error Correction for High-Frequency Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-frequency components in data cause significant restoration errors during irreversible encoding, leading to compromised data quality, especially in applications like display devices where calibration data compression is critical.
Innovation Solution
A data processing device incorporating an average correction processor, irreversible encoder/decoder, binary predictor, and error estimator to minimize restoration errors by predicting and correcting errors in encoded data, while maintaining efficient data compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If irreversible encoding is used to achieve high compression ratio, then data compression efficiency is improved, but restoration error increases
Solution Approach 1:
The encoding process is segmented into two distinct parts: an irreversible encoder for compression and a separate error correction module. The decoder is also segmented into an irreversible decoder and an error correction unit that uses the binary predictor and error estimator to separately handle compression and error correction functions.
Solution Approach 2:
A binary predictor acts as an intermediary between the irreversible decoder and the final output. It predicts whether restoration error exceeds a threshold and introduces an error estimator that mediates the correction process by estimating the magnitude of restoration error, allowing the system to selectively apply corrections only when needed.
2Measurement precision
If error correction is added to reduce restoration error, then data quality is improved, but device complexity increases
Solution Approach 1:
Instead of applying full error correction to all decoded data, the binary predictor determines whether correction is needed by comparing restoration error against a threshold. This partial action approach applies error correction only when necessary, reducing the overall complexity compared to universal correction schemes.
Solution Approach 2:
The system changes the parameter of error correction application from always-on to conditionally-applied based on the binary predictor's assessment. The error estimator dynamically adjusts correction magnitude based on estimated restoration error, optimizing the balance between quality improvement and complexity management.
3Measurement precision
If error correction data is stored to minimize restoration error, then data quality is improved, but encoded data size increases
Solution Approach 1:
Error correction data is stored only when the binary predictor determines that restoration error exceeds the threshold. This partial storage approach avoids the overhead of storing correction data for all encoded streams, minimizing the increase in encoded data size while still improving quality where needed.
Solution Approach 2:
The system applies different quality levels to different parts of the encoded data stream. High-quality error correction is applied only to segments where the binary predictor identifies significant restoration error, while other segments use standard decoding without additional correction overhead.
Data Source
AI summary
A data processing device includes an average correction processor that corrects input data based on an average value of the input data and outputs average correction data, and reversibly encodes and decodes the average value to generate a decoded average value, an irreversible encoder/decoder that encodes and decodes the average correction data and outputs first and second decoded data, a binary predictor that predicts a magnitude of a restoration error included in the second decoded data based on the first decoded data and the decoded average value and outputs a prediction result as binary data, an error estimator that outputs an estimation error based on the first decoded data, the second decoded data, and the decoded average value, and an operation circuit that outputs output data based on the second decoded data, the binary data, the estimation error, and the decoded average value.