Color Measurement Device Calibration Using Standard Raw Data Space
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing color measurement device calibration processes are complex, time-consuming, and require a large number of calibration samples, making them impractical for mass production.
Innovation Solution
A multi-step calibration process that generates and utilizes a collection of calibration functions to transform raw data from uncalibrated color measurement devices into a standard raw data space, allowing for calibration using fewer than 100 color samples, with a processor configured to derive transformation functions using a master calibration device and machine learning methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calibration methods using large training datasets (e.g., 7383 or 1950 color samples) are used to improve measurement precision, then the calibration accuracy is improved, but the time consumption and complexity of the calibration process increases significantly
Solution Approach 1:
The patent segments the calibration process into two distinct stages: (1) generating a standard raw data space dataset using a master calibration device with a large training set, and (2) deriving device-specific transformation functions using a small calibration set. This segmentation allows the time-consuming accurate calibration to be performed once to create reference data, while individual device calibration becomes fast and simple.
Solution Approach 2:
The patent performs preliminary action by pre-generating the standard raw data space dataset using a master calibration device with a comprehensive training set (7383 or 1950 samples). This pre-computed reference dataset is stored and reused for calibrating multiple devices, eliminating the need to repeatedly process large datasets for each device calibration.
2Measurement precision
If traditional calibration methods using large training datasets are used to improve measurement precision, then the calibration accuracy is improved, but the device complexity and operational complexity increases
Solution Approach 1:
The calibration process is segmented into distinct functional modules: (1) master device calibration to generate standard raw data space, (2) transformation function derivation using small calibration sets, and (3) application of device-specific transformation functions. This modular segmentation simplifies the overall process by breaking down the complex traditional calibration into manageable, reusable components.
Solution Approach 2:
The patent creates a virtual standard measurement device through the standard raw data space, which serves as a reference model for calibrating multiple physical devices. Instead of directly calibrating each device against physical standards, the system copies the standard measurement characteristics into a digital transformation function that can be applied consistently across devices.
3Measurement precision
If traditional calibration methods using large training datasets are used to improve measurement precision, then the calibration accuracy is improved, but the productivity and ease of manufacture decreases
Solution Approach 1:
The standard raw data space dataset is pre-computed using a master calibration device with a large training set, and this reference data is stored for reuse. This preliminary action enables rapid calibration of production devices without repeating the time-consuming large dataset processing, thereby improving mass production efficiency.
Solution Approach 2:
The patent changes the calibration approach from using large datasets for each device to using small calibration sets combined with pre-computed transformation functions. By transforming the calibration parameters from device-specific large-scale calibration to universal transformation functions derived from small sets, the system achieves both high accuracy and high productivity.
Data Source
AI summary
A system including a processor and a memory configured to store code executed by the processor is provided. In one or more implementations, the processor is configured by the code to calibrate measurements made by color measurement devices. In one particular implementation, the processor receives a measurement dataset of one or more color values, for a sample obtained by a color measurement device. The processor is configured to convert the measurement dataset to a standard space measurement dataset using a standard space measurement model and calculate a color dataset based on the standard space measurement dataset using a color conversion model. The processor is further configured to output the calculated color dataset to at least one of a display, database or local memory store.