Video Decoding Filtering Method for Blocking Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression techniques, such as H.264/AVC and HEVC, face inefficiencies in encoding and decoding due to increasing video size, resolution, and frame rate, necessitating improved encoding and decoding methods to enhance image quality and reduce hardware resource usage.
Innovation Solution
A filtering method that determines block boundaries for filtering in a reconstructed image, sets boundary strength based on block type and preset conditions, and applies filtering accordingly, using a video decoding apparatus with a boundary determination unit, strength setting unit, and filtering execution unit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If video compression techniques (H.264/AVC, HEVC) are used to reduce data volume, then storage and transmission efficiency improve, but blocking artifacts occur at block boundaries reducing image quality
Solution Approach 1:
The patent converts the harmful blocking artifacts caused by block-based compression into beneficial filtering opportunities. By detecting block boundaries and applying targeted filtering operations, the method transforms the visible artifacts into smoothed transitions while preserving the compression efficiency benefits of block-based coding.
Solution Approach 2:
The patent applies filtering operations selectively at block boundaries rather than uniformly across the entire image. By identifying specific boundary locations and applying appropriate filter strengths based on local characteristics (intra/inter prediction modes, boundary strength values), the method removes artifacts only where needed while preserving image details in non-boundary regions.
2Object-affected harmful factors
If filtering is applied to remove blocking artifacts, then image quality improves, but encoding and decoding complexity increases
Solution Approach 1:
The patent segments the filtering process into distinct stages: boundary detection, boundary strength determination, and conditional filter application. By dividing the reconstructed image into blocks and processing only boundary regions rather than the entire image, the method reduces computational complexity while maintaining artifact removal effectiveness.
Solution Approach 2:
The patent applies filtering partially rather than universally - only at detected block boundaries and only when boundary strength thresholds are exceeded. This selective application reduces the total number of filtering operations required compared to full-image filtering, thereby lowering encoding and decoding complexity.
3Productivity
If block-based compression is used, then encoding efficiency improves, but boundary discontinuities occur between adjacent blocks
Solution Approach 1:
The patent introduces filtering operations as an intermediary process between block-based compression and final image output. The filter acts as a mediator that smooths the discontinuities created by block boundaries while preserving the compression efficiency of the block-based approach, effectively bridging the gap between compressed blocks.
Solution Approach 2:
The patent applies filtering as a preliminary post-processing step after decomposition into blocks but before final image assembly. By addressing boundary discontinuities early in the reconstruction process, the method prevents artifact propagation and ensures smooth transitions before the image is finalized for display or further processing.
Data Source
AI summary
A filtering method comprises: determining at least one block boundary to which filtering is to be applied in a reconstructed image; setting a boundary strength of the block boundary based on prediction modes of two target luma blocks forming the block boundary among a plurality of prediction modes, a type of the block boundary and preset conditions; calculating an average value of quantization parameters which are respectively applied to the target luma blocks, and deriving a variable by adding the average value to an offset value derived based on offset information to be encoded at a level of a sequence parameter set that is a header which is referenced in common by the pictures belonging to the sequence; and determining whether to perform the filtering on the block boundary and performing the filtering based on the set boundary strength and the variable.