Spectral Characteristic Calculation Using Conversion Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional spectral diagnosis methods using electronic endoscopes require a spectrometer for accurate spectral characteristic measurement, which is not always feasible, and existing methods for estimating spectral reflectivity from RGB brightness values are less precise due to the difference in wavelength width between color components and the required narrow wavelength range.
Innovation Solution
A spectral characteristic calculation device and method that utilize a plurality of conversion matrices generated from sample data to evaluate the similarity of brightness information, selecting the most appropriate matrix for accurate spectral characteristic calculation based on Mahalanobis distance, allowing for precise estimation of spectral characteristics without a spectrometer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a spectrometer is used to measure spectral characteristics, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent creates a conversion matrix that copies the spectral measurement function from a spectrometer into a computational algorithm. By pre-calibrating conversion matrices from actual spectrometer measurements and storing them in a database, the system replicates spectrometer functionality through software calculation rather than physical hardware, thereby achieving spectral measurement precision without requiring a spectrometer in the final diagnostic device.
Solution Approach 2:
The patent replaces the mechanical/optical system of a spectrometer with an information processing system. Instead of using physical optical components to disperse and measure light spectra, the system uses RGB brightness values from a standard endoscope camera and applies pre-calculated conversion matrices through computational algorithms to derive spectral characteristics, substituting mechanical measurement with electronic calculation.
2Device complexity
If a single conversion matrix is used for spectral estimation, then device complexity is reduced, but measurement precision deteriorates when target differs from sample
Solution Approach 1:
The patent segments the single conversion matrix into multiple distinct conversion matrices, each stored as a separate entry in a database. Each conversion matrix corresponds to specific sample conditions (different tissues, illumination conditions, or spectral features). This segmentation allows the system to select the most appropriate conversion matrix for the specific diagnostic scenario, improving precision without significantly increasing device complexity since the matrices are stored data rather than physical components.
Solution Approach 2:
The patent changes the parameter of conversion matrix selection based on the specific characteristics of the target tissue or diagnostic requirements. By maintaining multiple conversion matrices with different calibration parameters and selecting the appropriate one based on the diagnostic context, the system adapts to different measurement conditions, thereby improving spectral estimation precision across various tissue types while keeping the device structure relatively simple.
Data Source
AI summary
A spectral characteristic calculation device stores a plurality of conversion matrices used to calculate the spectral characteristic values based on brightness values of a digital color image, and a plurality of pieces of brightness value information respectively corresponding to a plurality of sample groups. The plurality of conversion matrices are generated based on the sample color image information contained in the plurality of sample groups, respectively. An evaluating unit is configured to evaluate similarity of brightness value information of the target data with respect to each of the plurality of pieces of brightness value information of the plurality of sample groups. A conversion matrix corresponding to the sample group that is evaluated to have the highest similarity is selected as the conversion matrix. Then, a calculating unit calculates the spectral characteristic values of the target data using the conversion matrix selected by the selecting unit.