Fundus Image Color Tone Correction via Histogram Target Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fundus imaging using single-color light results in significant differences in color tone due to varying reflection characteristics, and is affected by alignment and focus deviations, making it difficult to stabilize color tone in fundus images.
Innovation Solution
An ophthalmologic image processing method and apparatus that corrects the color tone of fundus images by adjusting the gradation values of pixels to match a target pattern, using a computer to analyze and correct the histogram distribution of each channel of the fundus image, ensuring consistent color representation regardless of individual differences and imaging conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a beam of single-color light is used as illumination light, then the apparatus can be simplified and manufacturing cost reduced, but significant differences in color tone occur in the fundus image for every time imaging
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing correction values in a lookup table before actual imaging. The correction values are computed based on the relationship between wavelength, focus position, and color tone, then stored for rapid retrieval during imaging. This eliminates the need for complex real-time calculations while maintaining color tone consistency across multiple images.
Solution Approach 2:
The patent implements feedback by using the measured or predetermined color tone characteristics of single-color light imaging to generate correction values. These correction values are then applied to subsequent images to compensate for color tone variations. The system continuously refines the correction based on the relationship between focus position, wavelength, and observed color tone.
2Productivity
If alignment deviation and focus deviation occur during imaging, then imaging speed may be maintained, but color tone in the fundus image is significantly affected
Solution Approach 1:
The patent pre-calculates correction values for various focus positions and wavelengths before imaging. When imaging occurs with focus or alignment deviations, the system quickly retrieves the appropriate correction values from the pre-computed lookup table based on the actual focus position and wavelength used, enabling rapid correction without sacrificing imaging speed.
Solution Approach 2:
The patent changes parameters by adjusting the correction values based on the relationship between wavelength, focus position, and color tone. The system dynamically selects correction values corresponding to the actual imaging conditions (wavelength and focus position), allowing compensation for deviations without requiring re-imaging or complex real-time calculations.
3Reliability
If multi-color light (white light) is used as illumination light, then color tone in fundus images is stable, but the apparatus complexity and cost increase
Solution Approach 1:
The patent creates a computational copy of the color tone correction process. Instead of using complex hardware (multi-color light sources and multiple optical paths), the system creates a software-based correction model that replicates the color tone stabilization effect. The lookup table stores correction values that copy the effect of multi-color light imaging without requiring the corresponding hardware complexity.
Solution Approach 2:
The patent replaces the mechanical/optical system (multi-color light source and complex optical path) with a computational system. Instead of using white light and managing multiple wavelengths through optical components, the system uses single-color light with software-based correction algorithms that calculate and apply appropriate color tone adjustments based on pre-computed values.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method effectively suppresses color tone variations in fundus images, providing consistent and stable color representation, even with confocal type imaging systems, and allows for software-based corrections without hardware limitations.
Implementation Method 1
irradiating a fundus with a plurality of beams of single-color light having different wavelengths to capture a fundus color image based on fundus reflection light
Data Source
AI summary
An image processor performs a histogram acquisition step of acquiring a histogram representing a distribution of gradation values of pixels in a fundus color image captured by irradiating a fundus with a plurality of beams of single-color light having different wavelengths, the histogram being acquired for each channel corresponding to each beam of single-color light, a histogram correction step of acquiring a corrected histogram by correcting the histogram of each channel acquired in the histogram acquisition step, of which a target pattern is set for each channel in advance, so as to fit to the corresponding target pattern, and a color tone corrected image generation step of generating a color tone corrected image, in which a distribution of gradation values for each channel is represented by the corrected histogram, based on the corrected histogram of each channel.