Signal Processing System for Accurate Spectral Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current signal processing systems for identifying subjects with specific spectral characteristics, such as blood vessels, face challenges in accurately separating image signals into wavelength areas and calculating weighting coefficients for reliable identification, often resulting in noise and error.
Innovation Solution
A signal processing system comprising a base vector acquisition unit, separation unit, system spectral characteristics acquisition unit, calculation unit, and output signal calculation unit, which separates color signals into wavelength areas and calculates weighting coefficients based on known spectral characteristics and illumination light, enabling accurate identification of subjects with specific spectral characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image signals are separated into wavelength areas using conventional methods, then identification of subjects with specific spectral characteristics can be performed, but noise and error increase reducing reliability
Solution Approach 1:
The patent applies preliminary action by pre-acquiring spectral characteristics of the image acquisition system and illumination light, and pre-calculating base vectors corresponding to different wavelength areas before actual image processing. This preparation enables more accurate and reliable separation of image signals into wavelength areas during actual operation, reducing noise and error in the identification process.
Solution Approach 2:
The patent changes parameters by calculating weighting coefficients that dynamically adjust the contribution of each wavelength area based on the specific subject being identified. By modifying the weighting parameters according to the target subject's spectral characteristics, the system achieves higher identification reliability while maintaining signal separation precision through optimized parameter selection.
2Measurement precision
If conventional signal processing is used to identify subjects, then processing can be performed with standard equipment, but accuracy in identifying both surface and deep-layer features is insufficient
Solution Approach 1:
The patent applies segmentation by dividing the broadband illumination spectrum into multiple wavelength areas (e.g., blue, green, red regions) and processing each wavelength area independently with dedicated base vectors. This segmentation enables accurate identification of both surface and deep-layer features by capturing depth-information from different wavelength penetrations, achieving high identification accuracy without requiring complex hardware modifications.
Solution Approach 2:
The patent introduces weighting coefficients as an intermediary element that mediates between the separated wavelength area signals and the final identification result. These coefficients optimize the contribution of each wavelength area based on the target subject characteristics, enabling accurate identification of multi-layer features while maintaining manageable processing complexity through a unified computational framework.
3Loss of information
If broadband illumination is used for image acquisition, then more spectral information is obtained, but separation into wavelength areas becomes more difficult and error-prone
Solution Approach 1:
The patent applies preliminary action by pre-acquiring and storing the spectral characteristics of the illumination light and image acquisition system, and pre-calculating base vectors for each wavelength area before actual image processing. This preparation enables reliable separation of broadband illumination signals into distinct wavelength areas by providing reference data that guides the decomposition process, maintaining separation reliability despite the complexity of broadband spectra.
Solution Approach 2:
The patent changes parameters by dynamically adjusting weighting coefficients that control the contribution of each wavelength area during signal separation. By optimizing these parameters based on the specific broadband illumination source and target subject, the system achieves reliable separation of spectral information while retaining complete spectral data from the broadband illumination for accurate identification.
Data Source
AI summary
A base vector acquisition unit acquires a base vector based on spectral characteristics of a subject to be an identification object with spectral characteristics known, and a system spectral characteristics acquisition unit acquires a spectral characteristics of an image acquisition system including spectral characteristics concerning a color imaging system provided for image acquisition of subjects including the subject to be identification object, and spectral characteristics concerning illumination light used in image acquisition of subjects by the color imaging system. A calculation unit calculates weighting coefficient concerning the base vector for each wavelength area, from image signal composed of a plurality of color signals obtained in the color imaging system, the base vector and spectral characteristics of the image acquisition system. An output signal calculation unit calculates an output signal as identification result of the subject to be identification object based on the weighting coefficients.