Decoder Bit Converter for High-Bit-Depth Video Signal Error Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Video signal decoding quality deteriorates when encoded data with nine or more bits is decoded using an 8-bit reference frame due to bit errors, leading to poor picture quality.

Innovation Solution

A decoder system that includes a parser to extract header and data information, an intra predictor to generate predicted pixel values based on neighboring blocks, and an inter predictor with a bit converter to convert pixel values of reference blocks to match the bit length of current blocks, using matrices to calculate additional bits and enhance prediction accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If encoded data with nine or more bits is decoded using an 8-bit reference frame, then decoding complexity is reduced, but picture quality deteriorates due to bit errors

Engineering Contradiction:
Improvedecoding complexityVSAvoidpicture quality
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent changes the bit depth parameter of the reference frame from 8-bit to 9-bit or higher, matching the encoded data bit depth. This parameter change resolves the contradiction by enabling accurate decoding of high-bit-depth encoded data while maintaining decoding process simplicity, thereby improving picture quality without significantly increasing decoding complexity.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If reference frame bit depth is increased to match encoded data bit depth, then picture quality improves, but memory requirements and processing complexity increase

Engineering Contradiction:
Improvepicture qualityVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent performs preliminary conversion of the reference frame from 8-bit to 9-bit or higher bit depth before the decoding process. By preparing the reference frame in advance with the correct bit depth, the system avoids complex real-time bit depth conversion during decoding, thereby improving picture quality while minimizing increases in processing complexity.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If reference frame bit depth is increased to match encoded data bit depth, then decoding accuracy improves, but memory requirements increase

Engineering Contradiction:
Improvedecoding accuracyVSAvoidmemory requirements
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent changes the bit depth parameter of the reference frame to match the encoded data, improving decoding accuracy. While this does increase memory requirements due to the higher bit depth, the increase is minimal (e.g., from 8-bit to 9-bit is only a 25% increase per pixel) and is justified by the significant improvement in decoding accuracy and picture quality.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10271059B2Decoder, decoding system including the decoder and method of operating the decoder
Publication Date: 2019.04.23 SAMSUNG ELECTRONICS CO LTD
  • US10271059B2 patent drawing
  • US10271059B2 patent drawing
  • US10271059B2 patent drawing

AI summary

A decoder includes a parser, an intra predictor, and an inter predictor. The parser parses a bit stream to extract header information and first through third data information, to determine a prediction mode for blocks in a current frame based on the header information, and to selectively output the first and second data information based on the prediction mode. The intra predictor generates a predicted value of a pixel of a current block based on a pixel value of a first reference block in the current frame based on the first data information. The inter predictor predicts a second reference block from a reference frame based on the second data information, converts a pixel value of the second reference block, and generates a predicted value of the pixel of the current block based on a converted pixel value.