Lens Cleanliness Detection via Pupil Differential Flat Field Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lens inspection methods, particularly visual inspection and existing machine vision-based methods, suffer from low sampling rates, accuracy, and inefficiency in detecting surface defects, which affect lens cleanliness and lead to the production of defective products.
Innovation Solution
A method utilizing differential flat-field correction of a pupil, involving a narrow-band multispectral uniform surface light source, a camera, and varying pupil aperture sizes to collect and process bright-field and dark-field image data, performing pupil differential calculations to enhance impurity detection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If visual inspection method is used, then labor intensity is reduced, but detection accuracy and sampling rate are insufficient
Solution Approach 1:
The patent replaces manual visual inspection with an automated machine vision system that uses optical components (lens, camera, light source) and image processing algorithms to detect surface defects. The system captures images of the lens surface and automatically analyzes them to identify impurities, thereby achieving both automation and high detection accuracy simultaneously.
Solution Approach 2:
The patent employs differential flat-field correction by capturing images at multiple different illumination levels (bright-field and dark-field conditions) and mathematically processing them to enhance contrast. This parameter change in illumination conditions allows the system to detect subtle surface defects that would be invisible under normal lighting, significantly improving detection accuracy while maintaining automation.
2Productivity
If machine vision-based inspection is used, then productivity is improved, but detection efficiency for impurities is insufficient
Solution Approach 1:
The patent segments the inspection process into multiple stages: capturing bright-field images, capturing dark-field images, performing flat-field correction, and detecting impurities. By dividing the inspection into these distinct segments and processing them sequentially, the system maintains high productivity while thoroughly detecting impurities through comprehensive multi-stage analysis.
Solution Approach 2:
The patent changes illumination parameters by capturing images under both bright-field and dark-field conditions, then applies differential flat-field correction through mathematical processing. This parameter transformation enhances the contrast and visibility of impurities, making them easier to detect while maintaining the high-speed automated inspection process.
3Device complexity
If bright-field illumination is used, then image acquisition is simplified, but signal-to-noise ratio is insufficient for impurity detection
Solution Approach 1:
The patent merges the advantages of both bright-field and dark-field illumination by capturing images under both conditions and then combining them through differential flat-field correction. The bright-field images provide good overall illumination and simplicity, while the dark-field images enhance contrast for impurity detection. The mathematical combination of both datasets produces images with high signal-to-noise ratio while maintaining relatively simple device complexity.
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
This method improves the detection efficiency of lens impurities by suppressing the influence of low relative illumination, enhancing signal-to-noise ratio, and providing clear, edge-enhanced images of impurities, thereby improving the accuracy and efficiency of lens cleanliness inspection.
Implementation Method 1
the light signal obtained of the surface of the product is converted into an electrical signal
Data Source
AI summary
A method for detecting lens cleanliness of a lens in a flat-field optical path, the flat-field optical path includes a light source, the lens, a camera, the light source is a narrow-band multispectral uniform surface light source, the camera's light-sensitive surface is disposed perpendicular to an optical axis of the lens and in the light position of the lens and a pupil interposed between the lens and the light source, the method including collecting the bright-field image data and dark-field image data for a plurality of pupil aperture sizes through the lens; for each pixel, performing a pupil differential flat field correction to yield a plurality of PiPj pupil differentials; and displaying the pupil differentials in the form of a plurality of images to show uniformity of each the image, wherein a non-uniform area on each the image is determined to have been caused by an impurity of the lens.