Video Acuity Measurement Using Symbol Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for quantifying the performance of imaging systems do not effectively relate to human perception, making it difficult for users to select suitable video systems for specific applications, as they rely on theoretical models with little regard for human visual perception.
Innovation Solution
A method to measure video acuity by using a chart with test symbols of varying sizes, capturing digital images, cropping and aligning template images with test images, determining normalized correlation, and identifying the symbol size above which more than a predetermined fraction is correctly identified, to calculate video acuity in units of symbols per degree.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional theoretical models (MTF, sinusoidal cycles) are used to measure imaging system performance, then measurement precision is improved, but the relevance to human perception deteriorates
Solution Approach 1:
The patent uses copied human visual perception characteristics by employing test symbols (letters, numbers, shapes) that mirror what humans actually perceive in images. Instead of abstract sinusoidal patterns, the system uses recognizable symbols that replicate real-world visual tasks, making the measurement both precise and perceptually relevant
Solution Approach 2:
The patent changes the measurement parameter from abstract optical frequency (MTF) to perceptually meaningful symbol identification. By measuring the smallest clearly identifiable symbol size rather than sinusoidal cycle frequency, the system transforms the measurement parameter to bridge the gap between optical precision and human perception
2Manufacturing precision
If resolution targets with letters and line patterns are used, then manufacturing precision is improved, but ease of operation deteriorates due to complexity
Solution Approach 1:
The patent extracts only the essential element needed for perception measurement - clearly identifiable symbols - from complex traditional test charts. By removing unnecessary line patterns and complex geometric shapes, the system simplifies the test target to just what is needed for meaningful visual acuity assessment
Solution Approach 2:
Instead of having users manually analyze complex patterns and calculate resolution metrics, the patent inverts the approach by using automated computer vision algorithms to identify symbols and calculate acuity. This transforms a complex manual operation into a simple automated process
Data Source
AI summary
A method of measuring the video acuity of a physical imaging system is disclosed. A chart comprising an image of a set of test symbols is provided. The set of test symbols comprises a plurality of symbols repeated at a plurality of symbol sizes. A digital image of the chart is obtained using the imaging system. The digital image is cropped to obtain a set of cropped images comprising individual digital images of each test symbol in the set of test symbols. A template image of a test symbol is aligned with a test image comprising one member of the set of cropped images. The template image comprises a digitally generated image of the test symbol in the test image. The normalized correlation between the aligned template image and the test image is determined. The test symbol whose template image has the highest correlation with the test image is identified.