Saturated Pixel Detection Using Illuminant-Adaptive Color Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting over-exposed regions in images using a fixed saturation threshold can lead to incorrect detection due to varying light exposure across different color channels in camera sensors, particularly when the scene illumination is not uniform.
Innovation Solution
A method that adjusts saturation thresholds for each color channel based on the illuminant's color coordinates, either by using them directly or scaling them to maintain constant ratios, to accurately identify over-exposed pixels and prevent wrong detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed saturation threshold is used for all color channels, then the detection method is simple and fast, but the detection accuracy deteriorates due to varying light exposure across different color channels
Solution Approach 1:
The patent applies local quality by using different saturation thresholds for each color channel (R, G, B) instead of a single fixed threshold. Each channel has its own threshold value that adapts to the specific lighting conditions and sensor characteristics of that channel, thereby improving detection accuracy while accounting for the varying light exposure across different color channels.
Solution Approach 2:
The patent changes the parameter of saturation threshold from a fixed value to a variable value that depends on the illuminant color coordinates. The thresholds are dynamically adjusted based on the estimated illuminant characteristics, allowing the detection method to adapt to different lighting conditions and maintain high accuracy across various scenes.
2Reliability
If a fixed saturation threshold is used for all color channels, then the processing is fast and efficient, but wrong detection occurs when scene illumination is not uniform
Solution Approach 1:
The patent performs preliminary action by first estimating the illuminant color coordinates of the scene before determining the saturation thresholds. This preliminary estimation allows the subsequent threshold selection to be optimized for the specific lighting conditions, improving detection reliability by accounting for non-uniform illumination before the actual saturation detection is performed.
Solution Approach 2:
The patent dynamically changes the saturation threshold parameters based on the estimated illuminant characteristics. By adapting the thresholds to the specific scene illumination, the method maintains high detection reliability across different lighting conditions while avoiding wrong detections that would occur with fixed thresholds.
3Measurement precision
If saturation thresholds are adjusted based on illuminant color coordinates, then detection accuracy improves, but the computational complexity increases
Solution Approach 1:
The patent performs illuminant estimation as a preliminary step before saturation detection. By estimating the illuminant color coordinates first, the method establishes a foundation for selecting appropriate saturation thresholds without requiring complex real-time calculations during the actual detection phase, thereby balancing precision with computational efficiency.
Solution Approach 2:
The patent changes the saturation threshold parameters based on illuminant color coordinates to improve detection precision. This parameter adaptation allows the system to account for varying lighting conditions and sensor characteristics, achieving more accurate saturation detection while the computational overhead is managed through efficient illuminant estimation algorithms.
Data Source
AI summary
According to the invention, the saturation thresholds (thr; thg; thb) used for this detection depend on color coordinates (rw, gw, bw) representing an illuminant (ILL) of the image.