Adaptive MPEG Noise Reducer for Blocking Artifacts and Mosquito Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
MPEG compression artifacts such as blocking artifacts and mosquito noise in digital video signals are not adequately addressed by existing techniques, particularly in larger displays where noise is more noticeable, and current methods only provide marginal improvement in video quality.
Innovation Solution
An adaptive MPEG noise reduction system that includes detectors for blocking artifacts and mosquito noise, using customized filters to scan and remove noise from specific areas of the video signal without affecting image quality, employing buffering techniques and edge-adaptive filtering to identify and reduce noise based on its nature and location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If MPEG compression is used to reduce storage space and bandwidth requirements, then storage efficiency and transmission efficiency are improved, but compression artifacts such as blocking artifacts and mosquito noise are introduced
Solution Approach 1:
The video signal is divided into macroblocks and further into 4x4 pixel blocks for analysis. The noise reduction process segments the image into different regions (smooth areas, edge areas, textured areas) and applies different filtering strategies to each segment, allowing effective artifact removal while preserving important image features
Solution Approach 2:
Different filtering operations are applied to different regions of the image based on local characteristics. Smooth regions receive stronger filtering to remove artifacts, while edge and textured regions receive minimal or no filtering to preserve detail. The filtering strength and type are adapted locally to each block's content
2Object-generated harmful factors
If traditional noise reduction techniques are applied to reduce compression artifacts, then some noise is reduced, but image quality deteriorates due to loss of detail and introduction of blurring
Solution Approach 1:
The system analyzes each 4x4 block's variance and gradient characteristics to determine its type (smooth, edge, or textured). Different filtering operations are then applied: strong filtering for smooth blocks, selective filtering for edge blocks, and minimal filtering for textured blocks. This local adaptation preserves image quality while reducing artifacts
Solution Approach 2:
The filtering operation is dynamic and adaptive rather than static. The filter type and strength are determined on-the-fly based on local image characteristics. The system dynamically adjusts between different filtering modes (identity filter, horizontal filter, vertical filter, diagonal filter) based on the block's content and orientation
3Object-generated harmful factors
If filtering is applied to reduce mosquito noise and blocking artifacts, then noise is reduced, but edge sharpness and detail are lost
Solution Approach 1:
Edge detection is performed by calculating gradients in horizontal and vertical directions. Blocks identified as containing edges are treated differently from smooth blocks. For edge blocks, the system determines the edge orientation and applies filtering only perpendicular to the edge direction, preserving edge sharpness while removing artifacts in smooth regions
Solution Approach 2:
The system dynamically selects from multiple filter types including identity filter (no filtering), horizontal filter, vertical filter, and diagonal filter based on the block's characteristics. This dynamic selection allows the system to adapt to different local features and maintain edge sharpness while removing mosquito noise
4Object-generated harmful factors
If stronger filtering is used to remove compression artifacts, then artifact reduction is improved, but processing complexity and computational load increase
Solution Approach 1:
The complex task of noise reduction is broken down into simpler sub-tasks by segmenting the image into 4x4 blocks and further categorizing each block as smooth, edge, or textured. This segmentation allows the use of simple, computationally efficient filter operations rather than complex global processing, reducing overall computational load while maintaining effectiveness
Data Source
AI summary
The disclosed technology provides a system and a method for adaptive MPEG noise reduction. In particular, the disclosed technology provides a system and a method for reducing blocking artifacts and mosquito noise in an MPEG video signal. An overall MPEG noise detector may be used to determine the presence of noise in one or more frames of a video signal. When a sufficient amount of noise is detected in the one or more frames of the video signal, portions of the video signal that contain noise may be located and filtered to reduce the amount of noise present in the video signal.