White-Balance Noise Estimation for Low-Light Image Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image processing technologies fail to consider the influence of white balance correction on noise-induced coloring in low luminance regions, leading to deteriorated image quality.

Innovation Solution

An image processing apparatus that estimates the luminance and color difference components of noise based on both common gain and color-specific white balance gains, determining a luminance and color difference range for applying color suppression processing to suppress noise-induced coloring.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If a large gain is used to enhance visibility in low light environment, then visibility is improved, but noise component increases and dominates in low luminance regions causing coloring and image quality deterioration

Engineering Contradiction:
ImprovevisibilityVSAvoidnoise component
Core Design Contradiction:
Illumination intensityVSObject-generated harmful factors

Solution Approach 1:

The patent applies different processing strategies to different regions of the image based on luminance characteristics. Low luminance regions are identified and processed differently from high luminance regions, with specific noise suppression processing applied only where needed. This localized approach allows gain to be increased for visibility while preventing noise dominance in specific problematic regions.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent dynamically adjusts processing parameters based on luminance levels and noise characteristics. By changing processing intensity, color suppression strength, and filtering parameters according to local luminance conditions, the system can enhance visibility in dark regions without allowing noise to dominate, thus resolving the contradiction between visibility improvement and noise suppression.

Inventive Principle:
Principle #35Parameter changes

2Object-generated harmful factors

If conventional noise reduction control based on gain is applied, then noise suppression is achieved, but the influence of white balance correction on coloring caused by noise is not considered leading to incomplete noise suppression

Engineering Contradiction:
Improvenoise componentVSAvoidnoise suppression effectiveness
Core Design Contradiction:
Object-generated harmful factorsVSReliability

Solution Approach 1:

The patent merges noise suppression processing with white balance correction processing into a unified processing pipeline. By combining these previously separate operations, the system considers both gain-related noise and white balance-related coloring effects simultaneously, achieving more complete and reliable noise suppression that accounts for all sources of coloring.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent implements feedback mechanisms where noise characteristics and white balance correction effects are continuously evaluated and used to adjust processing parameters. This feedback loop ensures that the noise suppression processing effectively counteracts coloring caused by both gain and white balance correction, improving overall suppression reliability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3716617B1Image processing apparatus, image capturing apparatus, image processing method, and storage medium
Publication Date: 2025.09.17 CANON KK
  • EP3716617B1 patent drawingFigure 1
  • EP3716617B1 patent drawingFigure 2
  • EP3716617B1 patent drawingFigure 3~4

AI summary

There is provided an image processing apparatus. Estimation means estimates a color difference component of noise included in a captured image that has undergone white balance correction. Determination means determines a strength of color suppression processing based on a size of the color difference component of the noise. Processing means applies the color suppression processing to the captured image at the determined strength.