Video Noise and Flicker Reduction via Wavelet Subband Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video processing techniques fail to effectively reduce noise and flicker in video sequences due to limitations in camera technology and processing pipeline modules, leading to artifacts like blocking, aliasing, and temporal flicker.

Innovation Solution

A method involving pixel-adaptive warped spatial and temporal transforms with adaptive thresholding is applied to sub-frames of video frames, using a current input frame and a past output frame to derive and combine processed sub-frames, reducing noise and flicker through spatial and temporal processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional video processing techniques are used, then processing simplicity is maintained, but noise and flicker reduction effectiveness deteriorates

Engineering Contradiction:
Improvenoise and flicker reduction effectivenessVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The video signal is segmented into multiple frequency subbands using wavelet transform. Each subband is processed independently with appropriate filtering and thresholding operations, allowing targeted noise and flicker reduction while preserving important signal characteristics in different frequency ranges

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from spatial domain processing to wavelet domain processing, adding a frequency dimension to the analysis. By representing the video signal in the wavelet domain, the method can separately address noise and flicker components that are mixed in the spatial domain, improving reduction effectiveness without excessive complexity

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

2Object-affected harmful factors

If aggressive noise filtering is applied, then noise reduction is improved, but temporal flicker artifacts worsen

Engineering Contradiction:
Improvenoise levelVSAvoidtemporal flicker artifacts
Core Design Contradiction:
Object-affected harmful factorsVSObject-generated harmful factors

Solution Approach 1:

Different processing strategies are applied to different frequency subbands. Low-frequency subbands undergo temporal filtering to reduce flicker, while high-frequency subbands use thresholding to remove noise. This localized approach to each subband prevents the generation of temporal flicker artifacts while maintaining effective noise reduction

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent employs temporal filtering across multiple frames to continuously reduce flicker artifacts. By processing a sequence of frames and applying temporal constraints, the method maintains continuous suppression of flicker without introducing discontinuities or artifacts, while simultaneously addressing noise through wavelet thresholding

Inventive Principle:
Principle #20Continuity of useful action

3Object-affected harmful factors

If spatial filtering is used, then spatial noise is reduced, but temporal information and sharp transitions deteriorate

Engineering Contradiction:
Improvespatial noiseVSAvoidtemporal information and sharp transitions
Core Design Contradiction:
Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The patent uses adaptive thresholding in the wavelet domain that dynamically adjusts to local signal characteristics. This allows the filter to be more aggressive in homogeneous regions (reducing spatial noise) while being more conservative near edges and transitions (preserving temporal information and sharp transitions), achieving noise reduction without information loss

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The method changes the processing parameters (threshold values, filter strengths) based on the local wavelet coefficients and signal statistics. By adapting parameters to local conditions, the filter effectively reduces spatial noise in smooth regions while preserving important temporal information and sharp transitions where the signal varies rapidly

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2243298B1Noise and/or flicker reduction in video sequences using spatial and temporal processing
Publication Date: 2021.10.06 NTT DOCOMO INC
  • EP2243298B1 patent drawingFigure 1A~1B
  • EP2243298B1 patent drawingFigure 2
  • EP2243298B1 patent drawingFigure 3(A)~3(J)

AI summary

A method and apparatus is disclosed herein for reducing at least one of both flicker and noise in video sequences. In one embodiment, the method comprises receiving an input video and performing operations to reduce one or both of noise and flicker in the input video using spatial and temporal processing.