Fluorescence Image Correction Using Red Component Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fluorescence images captured by endoscopes often suffer from light-dark contrast issues due to varying imaging distances, making it difficult to distinguish between normal and abnormal tissues, and existing solutions like low-pass filtering or near-infrared imaging have limitations in maintaining signal-to-noise ratio and edge preservation.
Innovation Solution
An image processing device that captures normal light and fluorescence images using color filters with different spectral characteristics, extracts specific wavelength components, and generates a uniform image to correct fluorescence images, thereby improving contrast and maintaining signal-to-noise ratio by reducing edge effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a standardization image is generated based on reflected light from the observed region, then lightness and darkness variations caused by imaging distance are corrected, but edges in contrast regions (such as blood vessels) cause luminance values to affect the standardization process and decrease the S/N ratio
Solution Approach 1:
The patent extracts only the red component image from the normal light image, excluding other color components (blue and green) that contain edge information. This selective extraction removes the harmful edge effects while preserving the useful illumination distribution information needed for standardization.
Solution Approach 2:
The patent applies different processing to different color components by selecting only the red component for standardization, while ignoring blue and green components that contain edge information. This local differentiation allows the standardization process to focus on illumination variations without being contaminated by edge artifacts.
2Reliability
If low-pass filter process is applied to the standardization image, then edge effects are reduced and S/N ratio is improved, but the sharpness of contrast regions is lost
Solution Approach 1:
Instead of applying a low-pass filter that blurs all edges, the patent extracts only the red component which naturally has reduced edge information, while preserving the structural information in other color components. This selective approach maintains edge sharpness where needed.
3Measurement precision
If near-infrared light imaging is used to correct fluorescence images, then lightness and darkness variations are corrected, but additional light sources and complex processing circuits are required
Solution Approach 1:
The patent makes the existing normal light imaging unit serve dual purposes: capturing both color information for standardization and providing the red component specifically for fluorescence correction. This eliminates the need for separate near-infrared light sources and processing circuits.
Solution Approach 2:
The normal light image itself provides the correction data needed for fluorescence standardization through its red component, eliminating the need for external reference images or additional light sources. The system uses its own existing data to achieve the correction.
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 solution effectively corrects lightness and darkness variations in fluorescence images, enhancing the ability to differentiate between normal and abnormal tissues by improving contrast and maintaining signal-to-noise ratio without requiring additional light sources or complex processing circuits.
Implementation Method 1
a specific color filter, which is one of the plurality of color filters and transmits light in a wavelength band for which a light absorption property of a contrast region in the observed region is low
Implementation Method 2
a fluorescence image of the observed region, the fluorescence image being captured by a fluorescence imaging unit that receives fluorescence generated from the observed region irradiated with excitation light
Data Source
AI summary
An image processing device captures a normal light image of an observed region via a plurality of color filters having different spectral characteristics, respectively, generates a uniform image having low contrast by extracting each picture signal corresponding to a picture signal in a wavelength band for which light absorption characteristic of a contrast region in the observed region becomes low, and corrects each picture signal of a fluorescence image of the observed region by using the uniform image.