Video Codec Noise Reduction via Adaptive Quantization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital image and video compression techniques face inefficiencies in noise reduction, which affects compression quality and bandwidth usage, as noise is difficult to compress due to its random and unpredictable nature, often introduced by equipment, environment, and transmission sources.
Innovation Solution
A noise representation component generates a noise energy spectrum represented by transform coefficients, which are used to adaptively change dead-zones in quantization, suppressing noise during encoding and synthesizing noise during decoding, thereby improving compression efficiency and image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If traditional quantization is used without noise adaptation, then encoding complexity is low, but noise suppression effectiveness is poor
Solution Approach 1:
The noise energy spectrum is estimated and dead-zone parameters are determined before the actual quantization process. This preliminary noise analysis allows the quantization step to be adaptively adjusted for each coefficient based on local noise characteristics, improving noise suppression while maintaining computational efficiency through pre-calculation
Solution Approach 2:
The dead-zone parameter is dynamically changed based on the estimated noise energy spectrum. Instead of using a fixed quantization dead-zone, the system adapts the dead-zone size to local noise conditions, expanding dead-zones in high-noise regions and maintaining tighter quantization in low-noise regions, thereby optimizing noise suppression effectiveness
2Quantity of substance
If compression ratio is increased to reduce bandwidth, then transmission cost decreases, but image quality deteriorates
Solution Approach 1:
Different quantization dead-zones are applied to different frequency coefficients based on their local noise characteristics. Important low-frequency coefficients that carry image structure information use smaller dead-zones to preserve quality, while high-frequency coefficients in noisy regions use larger dead-zones for effective noise suppression, achieving local optimization of both quality and compression
Solution Approach 2:
The patent converts the harmful effect of noise into a beneficial compression opportunity. By estimating the noise energy spectrum, the system identifies regions where noise dominates signal content and applies aggressive quantization (large dead-zones) in those regions, effectively discarding noise while preserving important signal information, thereby improving compression efficiency without significant quality loss
Data Source
AI summary
Systems and methods for noise reduction are disclosed herein. The system includes a video codec system that can encode and/or decode video information. A noise representation component can identify flat regions and a quantizer can utilize the identified flat regions to suppress noise during compression. By suppressing noise during compression, the size of the video file to be compressed can be reduced, compression can use less resources and take less time, and the speed at which the compressed information is transferred can benefit. Noise can be reintroduced during the reconstruction of the video. Accordingly, both noise reduction and noise synthesis can be accomplished.