Ringing Artifact Reduction in Compressed Video Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for reducing ringing artifacts in compressed video signals often result in over-smoothing of details and poor edge preservation, especially when relying on small kernel sizes or simple edge detection schemes.
Innovation Solution
A method that classifies pixels based on gradient and absolute difference calculations, employs morphological operations for edge differentiation, and applies a 5×5 bilateral sigma filter for targeted low-pass filtering, adapting thresholds for effective ringing artifact reduction without relying heavily on MPEG information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If small kernel sizes (3×3) are used for filtering, then the algorithm is simple to implement, but ringing artifacts are not sufficiently removed
Solution Approach 1:
The patent implements a multi-scale filtering approach where 3×3 filtering is applied first, then 5×5 filtering is applied to regions that still contain ringing artifacts. This nested structure allows the simpler 3×3 filter to handle most cases while the larger 5×5 filter addresses remaining artifacts, resolving the contradiction between simplicity and effectiveness.
Solution Approach 2:
The patent segments the image into different regions based on edge detection results, applying different filtering strategies to different segments. Edge regions use one filtering approach while non-edge regions use another, allowing the system to achieve both simplicity in implementation and precision in artifact removal by treating different areas differently.
2Manufacturing precision
If large window sizes (15×15) are used for artifact removal, then ringing artifacts are better reduced, but image details are over-smoothed
Solution Approach 1:
The patent applies filtering operations locally rather than globally. By detecting edges and applying filters only in specific regions where ringing artifacts are present, the system achieves effective artifact removal without over-smoothing entire image regions. The filtering strength and kernel size are adapted to local image characteristics.
Solution Approach 2:
The patent applies filtering selectively rather than uniformly across the entire image. By identifying and targeting only the regions containing ringing artifacts for filtering, the system achieves sufficient artifact removal without applying excessive smoothing to the entire image, thus preserving image details while removing artifacts.
3Productivity
If simple edge detection schemes are used, then the algorithm is computationally efficient, but edge preservation is poor
Solution Approach 1:
The patent implements adaptive filtering where the filtering parameters and strength are dynamically adjusted based on local image characteristics such as edge strength and variance. This dynamic adaptation allows the system to preserve edges effectively while maintaining computational efficiency by applying stronger filtering only where necessary.
Solution Approach 2:
The patent uses feedback from edge detection results to control the filtering process. Edge detection outcomes are used to adjust filtering parameters and determine which regions require filtering, creating a feedback loop that improves edge preservation while maintaining computational efficiency through selective processing.
Data Source
AI summary
A method and apparatus for ringing artifacts reduction for compressed video signals. The method includes receiving luma data to the digital signal processor, calculating sum of gradient of the luma data; calculating SAD of the luma data; performing pixel classification based of the calculated SAD and sum of gradient, performing erosion on a detected edge pixel indicator on a detected flat pixel indicators, determining at least one of the strength or weakness of the an edge based on the determined edge erosion, performing horizontal dilation on the detected edge pixel indicators and edge strength; and performing at least one of sigma or bilateral filtering to the luma data according to the detected edge pixel indicator, flat pixel indicator, edge strength, the number of very flat pixel in the block.