Dominant Color Determination Using Spatial Weighting and Chroma Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for determining dominant colors in digital images rely on mathematical algorithms that average color values, often resulting in low chroma colors and emphasizing background colors, which differ from human perception.
Innovation Solution
A system that divides a digital image into pixel groups with varying sample rates, weighting pixels closer to the center more heavily, and excluding pixels with low chroma or lightness values, to determine a dominant color that better mimics human perception, and outputs a matching paint color.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mathematical algorithms average color values across all pixels, then processing is simple and fast, but the result emphasizes background colors and produces low chroma colors that differ from human perception
Solution Approach 1:
The image is divided into multiple regions (e.g., foreground region and background region) based on spatial position. Different sampling strategies are applied to different regions: higher sampling density for foreground pixels and lower sampling density for background pixels. This segmentation resolves the contradiction by enabling perceptually accurate color determination while maintaining processing efficiency through region-specific optimization.
Solution Approach 2:
Different quality levels of analysis are applied to different parts of the image. Foreground pixels, which are more important for human perception, receive more intensive analysis with higher sampling rates. Background pixels receive less intensive analysis with lower sampling rates. This local differentiation improves measurement precision for the most important regions while reducing overall computational complexity.
2Measurement precision
If all pixels are analyzed at high sample rate, then dominant color accuracy improves, but processing time and computational resources increase
Solution Approach 1:
Instead of analyzing all pixels at maximum detail, the system applies partial analysis: high-sample-rate analysis is applied only to foreground pixels that contribute most to perceived dominant color, while background pixels are analyzed at lower sample rates. This partial action approach maintains accuracy for the most important regions while significantly reducing total processing time and computational resource requirements.
3Area of stationary object
If background pixels are included in analysis, then complete image coverage is achieved, but the dominant color result is skewed toward background colors
Solution Approach 1:
The image is segmented into foreground and background regions with different weighting schemes. Foreground pixels are assigned higher weights in the color averaging process, while background pixels are assigned lower weights. This segmentation enables the system to maintain complete image coverage while preventing background colors from skewing the dominant color determination, thereby improving measurement precision to match human perception.
Solution Approach 2:
The system changes the parameter of pixel weighting based on spatial position. Pixels closer to the center of the image (typically foreground) are given higher weight values, while pixels farther from the center (typically background) are given lower weight values. This parameter change resolves the contradiction by maintaining full image coverage while ensuring that the dominant color reflects foreground prominence as perceived by humans.
Data Source
AI summary
Systems and methods for determining a dominant color in a digital image are provided and include selecting pixels a digital image and applying a chroma filter, a low lightness filter, and/or a high lightness filter to the selected pixels. The remaining pixels after filtering are grouped based on hue, chroma, and lightness. A dominant color for the digital image is determined based on a color associated with the group that has the greatest number of pixels.