Optical Lens Tinting Model Inversion for Target Color Reproduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for tinting optical lenses are time-consuming and expensive due to the difficulty in reproducing exact desired colors, and neural networks require extensive data and are difficult to validate, leading to potential errors.
Innovation Solution
A method involving a two-model approach using a first model based on the Beer-Lambert law and a second model with error functions to accurately determine the tinting process variables, allowing for efficient reproduction of desired lens colors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a very large palette of colors is proposed to the wearer, then the choice of tint is improved, but the difficulty of reproducing the exact desired color increases
Solution Approach 1:
The patent applies parameter changes by using a colorimetric model that mathematically relates process parameters (dye concentrations, processing conditions) to colorimetric parameters (L*a*b* values). This allows precise control and reproduction of any desired color within the palette by calculating the specific parameter values needed to achieve the target color, thereby maintaining both versatility and manufacturing precision.
Solution Approach 2:
The patent replaces the traditional trial-and-error mechanical approach with an automated colorimetric measurement and calculation system. A spectrophotometer measures the actual color of produced lenses, and a computer calculates the necessary parameter adjustments to achieve the target color, eliminating manual visual assessment and iterative remanufacturing.
2Extent of automation
If neural networks are used to determine coloring recipes, then the ability to learn optimal tints is improved, but the requirement for large amounts of data and validation difficulty increases
Solution Approach 1:
The patent uses a simpler, more transparent mathematical model (colorimetric model based on Beer-Lambert law) instead of complex neural networks. This model requires minimal training data and can be validated easily, while still achieving automated determination of optimal coloring recipes. The solution trades the high complexity of neural networks for a more accessible and verifiable approach.
3Manufacturing precision
If multiple tests are conducted to obtain the desired tint, then the accuracy of the target color is improved, but the time and cost increase
Solution Approach 1:
The patent implements a feedback loop where the actual color of the produced lens is measured using a spectrophotometer, compared to the target color, and the difference is used to calculate parameter adjustments for the next production attempt. This automated feedback system enables convergence to the target color with minimal iterations, reducing both time and cost while maintaining high accuracy.
Solution Approach 2:
The patent performs preliminary colorimetric measurements and calculations before full production to determine the optimal parameters. By pre-calculating the expected color based on the colorimetric model and adjusting parameters in advance, the system minimizes the need for multiple post-production tests, saving time and resources.
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
Enables accurate and efficient reproduction of desired lens colors with fewer samples, simplifying the process and reducing costs by fine-tuning the model to ensure accuracy.
Implementation Method 1
said first model is based on a linear combination of the values of the process variables, said values preferably relating to concentration of dyes. said first model is based on the Beer-Lambert law.
Data Source
Figure 1
AI summary
The invention relates to a method for obtaining an optical lens with a target tint comprising steps of: - providing a process for tinting the optical lens to the target tint and a set of process variables suitable to modify the target tint, each process variable being able to take a value, - providing several optical lens samples tinted according to said process for tinting and associated to a set of values for said process variables, - measuring a measured tint of each optical lens sample, - providing models receiving as an input a set of values for the process variables and giving as an output a modelled tint for an optical lens tinted by said process, - fine-tuning said model by minimizing of a first function that depends on a gap between the measured tints and second modelled tints for the optical lens samples, - inverting said model, - performing the process using the set of values to obtain the optical lens having the target tint.