Parallel Wavelet Decoding With Overlapping Coefficient Strips

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing parallel implementations of image processing algorithms with decorrelative transforms face inefficiencies due to synchronization points and cache misses between processing channels, leading to decreased performance.

Innovation Solution

A method for parallel decoding that organizes wavelet coefficients into spatially related groups, allowing each processing channel to reconstruct image strips independently using extended groups of coefficients that overlap to ensure all necessary data is available, reducing dependencies and cache misses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If computation tasks are distributed among a plurality of processing channels to improve processing speed, then productivity is improved, but dependencies and synchronization points are introduced between channels which worsen performance

Engineering Contradiction:
Improveprocessing speedVSAvoidsynchronization waiting time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The image data is segmented into independent strips that can be processed by different processing channels simultaneously. Each strip is assigned to a specific processing channel, and the wavelet coefficients are organized into groups that correspond to these strips, enabling independent parallel processing without inter-channel dependencies.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a new organizational dimension for wavelet coefficients by grouping them according to the strips they belong to, rather than processing them in the traditional sequential manner. This reorganization allows processing channels to operate independently on their assigned strips while maintaining data integrity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If processing channels are synchronized to ensure correct data processing, then reliability is improved, but processing speed deteriorates due to idle waiting time

Engineering Contradiction:
Improvedata processing correctnessVSAvoidprocessing speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The wavelet coefficients are pre-organized into groups that correspond to specific image strips before processing begins. This preliminary organization ensures that each processing channel has immediate access to all the coefficients it needs for its assigned strip, eliminating the need for synchronization waiting during the processing phase.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If processing channels have their own cache memory to improve access speed, then productivity is improved, but cache misses increase due to data dependencies between channels which worsen performance

Engineering Contradiction:
Improvedata access speedVSAvoidcache miss penalty
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

By segmenting the wavelet coefficients into groups that correspond to specific strips, the patent ensures that each processing channel operates on a self-contained data set. This segmentation minimizes the need for channels to access each other's cache memory, reducing cache misses and improving overall processing efficiency.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4712478A1Method and apparatus for parallel decoding of digital data
Publication Date: 2026.03.18 INTOPIX
  • EP4712478A1 patent drawingFigure 1~2
  • EP4712478A1 patent drawingFigure 3~4
  • EP4712478A1 patent drawingFigure 5

AI summary

The invention relates to a method for decoding an encoded image obtainable by applying at least one wavelet-type filter operation in a vertical direction, wherein the efficiency of the decoding is improved by reducing dependencies between respective processing channels decoding in parallel respective strips of spatial image samples, wherein each processing channel reconstructs a strip of spatial image samples independently from the other processing channels from an extended group of wavelet coefficients, whereby extended groups of wavelet coefficients mutually overlap to an extent that each extended group of wavelet coefficients comprises all the wavelet coefficients required for the processing channel receiving the extended group of wavelet coefficients to independently reconstruct a strip of spatial image samples. The invention also relates to an apparatus and a computer program for performing the method for decoding of the invention.