Camera Color Image Processing Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional camera color image processing techniques fail to accurately represent human vision by treating all colors uniformly, neglecting chromatic attributes and lighting conditions, resulting in images that do not accurately reflect real-world perceptions.

Innovation Solution

A computerized method for image processing that separates achromatic and chromatic colors based on color temperature and lux, applying distinct weight values to each to generate adjusted color appearances, mimicking human visual and cognitive processing, thereby improving image quality and user experience under various lighting conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional processing techniques treat all colors in the same manner using simple mathematical models, then the processing is simple and fast, but the color accuracy and representation of human vision deteriorates

Engineering Contradiction:
Improvecolor accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the color processing into distinct pathways: achromatic color processing and chromatic color processing. Each pathway applies different processing rules and parameters, allowing the system to handle different color types with appropriate complexity levels rather than using a single complex model for all colors.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing qualities to different color regions. Achromatic colors (grayscale regions) receive processing focused on luminance and color temperature, while chromatic colors receive processing that preserves hue and saturation. This local differentiation improves overall color accuracy without uniformly increasing complexity across all processing.

Inventive Principle:
Principle #3Local quality

2Reliability

If conventional techniques use uniform processing for all colors, then the processing is straightforward, but the representation of lighting conditions and color casts deteriorates

Engineering Contradiction:
Improverepresentation accuracyVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent changes processing parameters based on color type and lighting conditions. For achromatic colors, it adjusts luminance and color temperature parameters. For chromatic colors, it preserves hue and saturation parameters. This dynamic parameter adjustment improves representation accuracy of different lighting conditions without requiring a completely complex processing framework.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If the processing separates achromatic and chromatic colors with different weight values, then the color appearance accuracy improves, but the processing complexity increases

Engineering Contradiction:
Improvecolor appearance accuracyVSAvoidprocessing algorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments color processing into achromatic and chromatic pathways, applying different weight values and processing rules to each segment. This segmentation allows the system to achieve high color appearance accuracy for each type without requiring complex interactions between all color channels, thus managing algorithmic complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic processing where the system adapts its processing approach based on the detected color type and lighting conditions. The processing weights and parameters are dynamically adjusted for achromatic versus chromatic regions, improving accuracy while keeping the overall algorithm structure manageable through conditional logic rather than universal complexity.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3942790B1Camera color image processing
Publication Date: 2024.04.24 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3942790B1 patent drawingFigure 1
  • EP3942790B1 patent drawingFigure 2
  • EP3942790B1 patent drawingFigure 3

AI summary

An image processing system receives image data acquired by an imaging device, and separates one or more achromatic colors from one or more chromatic colors in the received image data. The achromatic colors and the chromatic colors are processed based at least on a color temperature and lux to generate an adjusted color appearance for the achromatic colors by applying one or more defined weight values associated with the achromatic colors or the chromatic colors, respectively. The defined weight values associated with the one or more achromatic colors are different than the defined weight values associated with the one or more chromatic colors. A final image is generated using the adjusted color appearance for the one or more achromatic colors and the adjusted color appearance for the one or more achromatic colors that is more aligned with human vision and cognitive systems.