Imaging System Spectral Classification via Luminance Pattern Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current classification processes in factory automation and medical fields require significant processing load due to the use of hyperspectral image data, which can be cumbersome and inefficient, especially when classifying subjects by type using spectral information.
Innovation Solution
An imaging system featuring a filter array with different transmission spectra, an image sensor, and a processing circuit that generates luminance pattern data to classify subjects by type without the need for hyperspectral image data, utilizing compressed image data and a reconstruction table to reduce processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hyperspectral image data is used for subject classification, then spectral information and shape information are obtained, but processing load increases significantly
Solution Approach 1:
The patent extracts only the necessary spectral information features from the full hyperspectral data cube, rather than processing all spectral bands. By identifying and extracting key spectral signatures that are sufficient for classification, the system maintains measurement precision while significantly reducing the volume of data that requires intensive processing.
Solution Approach 2:
The classification process is segmented into multiple stages: initial subject detection, spectral feature extraction, and classification. By dividing the processing into discrete segments, the system can apply different processing intensities to different stages, reducing overall computational load while maintaining accuracy where it matters most.
2Measurement precision
If hyperspectral image data is used for subject classification, then spectral information is obtained, but processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-identifying and storing spectral signature templates for known subjects before actual classification occurs. During runtime, instead of analyzing complete hyperspectral data, the system quickly compares captured spectral data against pre-computed templates, dramatically reducing processing time while maintaining classification accuracy.
Solution Approach 2:
The patent applies partial action by processing only a subset of spectral bands that contain the most discriminative information for classification. Rather than analyzing the entire spectral range, the system identifies and processes only the critical wavelength regions, reducing processing time while preserving measurement precision.
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
This approach significantly reduces the processing load for subject classification by type, allowing for high sensitivity and spatial resolution without the need for hyperspectral image data generation, thereby enhancing efficiency and accuracy.
Implementation Method 1
a filter array that includes filters having different transmission spectra
Implementation Method 2
an image sensor that images light passing through the filter array and generates image data
Data Source
AI summary
There is provided an imaging system including: a filter array that includes filters having different transmission spectra; an image sensor that images light passing through the filter array and generates image data; and a processing circuit, in which the processing circuit acquires luminance pattern data generated on the basis of subject data that includes spectral information of at least one substance, the luminance pattern data being generated by predicting a luminance pattern detected when the substance is imaged by the image sensor, acquires first image data obtained by imaging a target scene by the image sensor, and generates output data regarding whether the substance is present in the target scene by comparing the luminance pattern data with the first image data.