Optical Aid Optimization via Machine Learning Visual Acuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining subjective visual acuity are laborious, prone to disruptive influences, and lack accuracy due to the complexity of neural transfer functions and the need for subjective test subject measurements.
Innovation Solution
A method utilizing machine learning to automatically determine subjective visual acuity by training an artificial neural network with a training data set, parameterizing stimulus images, and iteratively adapting parameters to optimize optical aids for improved visual acuity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subjective test subject measurements are used to determine visual acuity, then the measurement can be performed, but the process becomes laborious and susceptible to disruptive influences such as fatigue
Solution Approach 1:
The patent replaces the mechanical/manual process of subjective test subject measurements with an automated image processing system using machine learning. The neural network automatically analyzes stimulus images and determines visual acuity thresholds without requiring continuous human observer input, thereby reducing measurement time and eliminating fatigue-related errors while maintaining measurement accuracy.
2Measurement precision
If neural transfer functions are used to model the visual system, then visual acuity can be determined, but the functions are highly non-linear and difficult to describe, requiring adjustable weightings that must be adapted
Solution Approach 1:
The patent transforms the complex non-linear neural transfer function problem into a more manageable form by using the neural network to directly process stimulus images and determine visual acuity thresholds. The system learns optimal parameter configurations during training and automatically adapts to individual visual characteristics without requiring manual adjustment of complex weighting parameters, thereby reducing model complexity while maintaining measurement precision.
3Measurement precision
If multiple examinations are conducted to optimize optical aids, then sufficient accuracy can be achieved, but the process becomes laborious and time-consuming
Solution Approach 1:
The patent performs preliminary training of the neural network on a comprehensive dataset of stimulus images with various optical properties before actual optimization. This preliminary action allows the system to pre-learn the relationships between optical aid properties and visual acuity outcomes, enabling rapid prediction and optimization of optical aids without requiring multiple time-consuming examinations, thereby improving productivity while maintaining optimization accuracy.
Data Source
AI summary
A method for optimizing an optical aid by way of automatic measurement of the subjective visual performance, a method for producing a correspondingly optimized optical aid, an apparatus for producing optical aids, a computer program having a program code for carrying out the optimization method, which program can be run on a processor, and a non-transitory storage medium comprising the computer program stored thereon are disclosed. The method for optimizing the optical aid includes the automatic determination of subjective visual acuity using machine learning.