MTF Evaluation for Red/Clear Filtered Optical Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing modulation transfer function (MTF) evaluation methods for camera systems with color filters, such as Red/Clear (RCCC) filters, face challenges in accurately interpolating values for filtered pixels, as they assume monochrome images, leading to inefficiencies and inaccuracies in MTF calculations.
Innovation Solution
A system and method that utilize a slanted-edge technique with a Red/Clear color filter, where a pattern source provides an image with filtered and unfiltered pixels, and an image analysis system determines the axis of lowest curvature to generate an interpolated image, allowing for accurate MTF determination by selecting brightness values from a Moore neighborhood of pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If existing MTF evaluation methods assume monochrome images for color filters, then the interpolation process becomes simpler, but the accuracy of MTF calculations deteriorates
Solution Approach 1:
The patent applies local quality by treating filtered and unfiltered pixels differently in the interpolation process. Instead of uniformly processing all pixels as in monochrome images, the method selectively identifies filtered pixels (requiring interpolation) versus unfiltered pixels (providing reference values), and applies interpolation only where necessary based on local pixel characteristics and the curvature of the slanted edge in different directions.
2Productivity
If conventional slanted-edge methods are used with RCCC filters, then the evaluation process is faster, but the alignment sensitivity increases leading to inaccuracies
Solution Approach 1:
The patent implements dynamics by adaptively determining the optimal interpolation direction for each filtered pixel based on the local curvature characteristics of the slanted edge. Rather than using a fixed interpolation approach, the method dynamically selects between horizontal, vertical, or diagonal interpolation directions by calculating curvature values, thereby adapting to the local geometry and reducing alignment sensitivity while maintaining evaluation speed.
3Ease of operation
If standard interpolation is applied to all pixels in color filter arrays, then the processing is more uniform, but the MTF evaluation accuracy beyond Nyquist frequency deteriorates
Solution Approach 1:
The patent applies segmentation by dividing the pixel array into filtered pixels (requiring interpolation) and unfiltered pixels (providing reference data), and further segmenting the interpolation approach by direction (horizontal, vertical, diagonal). This segmented strategy allows selective application of interpolation only where needed and only in the optimal direction, improving accuracy beyond Nyquist frequency while maintaining operational simplicity through automated direction selection.
Data Source
AI summary
Systems and methods are provided for evaluating a modulation transfer function (MTF) of a device under test comprising a red/clear color filter. A pattern source provides a pattern. The device under test images the pattern to provide an image comprising a plurality of filtered pixels and a plurality of unfiltered pixels, each having an associated brightness value. An image analysis system determines, for each filtered pixel, an axis, of a plurality of axes, having a lowest curvature. A first axis of the plurality of axes is not orthogonal to a second axis. An interpolated image is generated having a new brightness value for each of the plurality of filtered pixels as a function of respective brightness values for a set of unfiltered pixels selected from a first-order Moore neighborhood of the pixel along the axis having the lowest curvature. The modulation transfer function is determined from the interpolated image.