Spectral Camera Calibration via Luminance Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing calibration apparatuses for spectral cameras face challenges in accurately correcting for wavelength unevenness and chromaticity, especially when measuring dark targets, due to fixed light source intensity and limited color conversion accuracy.
Innovation Solution
A calibration apparatus that includes processors to normalize measurement values by luminance, calculate correction matrices based on normalized values, and perform exposure correction, using a light source with an integrating sphere to uniformize light and output multiple colors, including black and low-gradation colors, to improve color conversion accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the light source intensity is fixed and correction matrix elements are calculated using least-squares method, then the calibration process is simple, but chromaticity accuracy deteriorates when measurement target is dark
Solution Approach 1:
The patent applies dynamics by making the light source intensity variable rather than fixed. The light source is controlled to output multiple luminance values, enabling the system to adapt calibration conditions to different brightness levels. This resolves the contradiction by allowing simple least-squares calculation for each luminance level while improving dark target chromaticity accuracy through multi-luminance data aggregation.
Solution Approach 2:
The patent changes the luminance parameter of the light source to multiple discrete values during calibration. By capturing spectrum images at different luminance levels and aggregating the correction matrix elements, the system achieves both operational simplicity (automated multi-luminance calibration) and improved measurement precision (reduced conversion errors for dark targets).
2Manufacturing precision
If correction matrix is calculated from reflectance of each wavelength, then wavelength unevenness is corrected, but conversion errors remain approximately the same regardless of brightness
Solution Approach 1:
The patent introduces dynamic luminance variation into the calibration process. Instead of using fixed light source intensity, the system varies luminance across multiple levels and aggregates correction results. This maintains wavelength unevenness correction capability while reducing brightness-dependent conversion errors through multi-luminance data synthesis.
Solution Approach 2:
The patent performs preliminary calibration actions at multiple luminance levels before final correction application. By pre-capturing spectrum images at various brightness levels and calculating correction matrix elements for each level, the system prepares comprehensive correction data that accounts for brightness variations, thereby reducing conversion errors across different target brightness conditions.
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 enhances color conversion accuracy for both bright and dark colors, reducing conversion errors and improving chromaticity determination, especially in dark environments, by normalizing measurement values and calculating correction matrices that account for varying light intensities.
Implementation Method 1
an integrating sphere 12 which uniformizes image light from the display apparatus 11 and emits the uniformized image light
Data Source
AI summary
There is provided a calibration apparatus including one or a plurality of first processors programmed to: obtain spectrum images from a spectral camera that images light from alight source portion; obtain a spectral reference value from a measurement result of a calibration reference device that measures the light; extract a gradation value at a correction point that is a pixel which generates a correction matrix among the spectrum images as a measurement value; divide the measurement value at the correction point and the spectral reference value by a luminance value of the light emitted from the light source portion to obtain a normalized measurement value and a normalized reference value; and calculate the correction matrix based on the normalized measurement value and the normalized reference value.