Endoscope System Using RGB Difference Values for Lesion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing endoscope systems face high processing loads and image evaluation value fluctuations due to non-linear calculations and brightness variations, making precise diagnosis of lesions challenging, especially for mild color differences.
Innovation Solution
An endoscope system that allocates pixel data to a plane defined by R and G (or B) components, calculating distance data along a third axis nonparallel to these components, and uses this data to standardize evaluation values, reducing processing load and brightness-induced fluctuations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If tone enhancement process and color space conversion are executed to enhance lesion detection accuracy, then measurement precision is improved, but device complexity and processing load increase
Solution Approach 1:
The patent extracts only the essential color information (R-G and R-B difference values) from the full RGB color data, discarding redundant information. This extraction approach maintains lesion detection capability while significantly reducing processing complexity by focusing only on the chromatic components that differentiate lesions from normal tissue.
Solution Approach 2:
Instead of applying complex tone enhancement and color space conversion to the entire image, the patent inverts the approach by directly calculating simple difference values (R-G and R-B) that inherently highlight color variations. This inversion simplifies the processing while maintaining or improving lesion detection accuracy.
2Measurement precision
If tone enhancement process is executed to improve lesion visibility, then measurement precision is improved, but the evaluation value fluctuates depending on brightness conditions
Solution Approach 1:
The patent applies local quality by calculating difference values (R-G and R-B) that are inherently insensitive to overall brightness changes. By focusing on relative color differences rather than absolute intensity values, the method maintains stable evaluation across varying illumination conditions while still detecting subtle color variations indicative of lesions.
3Measurement precision
If complex color space conversion is performed to calculate lesion index, then measurement precision is improved, but productivity decreases due to heavy processing load
Solution Approach 1:
The patent uses simple, computationally inexpensive calculations (basic subtraction operations for R-G and R-B differences) instead of complex color space conversions. These simple calculations can be performed rapidly on standard hardware, enabling real-time lesion detection without requiring expensive specialized processing equipment.
Data Source
AI summary
This endoscope system comprises: an image acquiring means for acquiring a color image having at least three color components; a placement means for placing points corresponding to respective pixels forming the color image in accordance with the color components of the points, within a plane including a first axis that is an axis of a first component among the at least three color components and a second axis that is an axis of a second component among the at least three color components and that intersects with the first axis; a distance data calculating means for calculating data regarding the distances between the points corresponding to respective pixels and a third axis defined as an axis that passes through a point at which the first axis and the second axis intersect with each other in the plane and that is not parallel to the first axis and the second axis; and an evaluation value calculating means for calculating a certain evaluation value for the color image on the basis of the calculated distance data.