Sub-pixel Defect Size Measurement via Spectral Diffuse Reflection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect detecting devices are limited in detecting defects smaller than a pixel unit, as they cannot measure the size of defects smaller than the pixel size, despite being able to detect their position through abnormality in pixel signal values.
Innovation Solution
A defect detecting device that uses an illumination unit emitting light with multiple spectral wavelengths, an imaging unit capable of capturing spectral images, and a detection unit that performs edge detection and generates a composite image to determine the size of defects by identifying diffuse reflection regions, allowing for the detection and measurement of defects smaller than a pixel unit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional imaging units are used to detect defects, then defect position can be detected through pixel signal abnormality, but the size of defects smaller than a pixel unit cannot be measured
Solution Approach 1:
The patent introduces spectral wavelength as an additional dimension beyond spatial pixels. By capturing multiple spectral images at different wavelengths and analyzing the spectral characteristics of diffuse reflection, the system achieves sub-pixel defect size measurement. The spectral dimension allows differentiation of defect sizes based on wavelength-dependent reflection characteristics, enabling measurement precision beyond the spatial pixel limit.
Solution Approach 2:
The patent changes the detection parameter from spatial pixel intensity to spectral wavelength characteristics. By analyzing how diffuse reflection varies across different spectral wavelengths, the system can quantify defect sizes. The spectral image data provides a wavelength-dependent parameter that correlates with defect dimensions, enabling precise measurement without increasing spatial resolution hardware.
2Measurement precision
If spectral images at multiple wavelengths are captured to measure defect size, then defect size smaller than pixel unit can be measured, but the device complexity and imaging process become more complex
Solution Approach 1:
The imaging unit is designed to capture multiple spectral images using a unified system that can operate at different wavelengths. The same imaging hardware captures spectral data across multiple wavelengths, and the detection unit processes all wavelengths through a consistent edge detection and spectral analysis algorithm. This multi-functional approach avoids requiring separate specialized devices for each wavelength, managing complexity while achieving comprehensive spectral measurement.
3Measurement precision
If edge detection is performed on each spectral image to detect diffuse reflection regions, then defect detection accuracy improves, but processing time and computational complexity increase
Solution Approach 1:
The system performs edge detection on each spectral image as a preliminary step to identify potential diffuse reflection regions. By pre-processing each wavelength image to extract edge information before spectral analysis, the system efficiently identifies candidate defect locations. This preliminary edge detection guides subsequent spectral wavelength analysis, reducing the computational burden of analyzing all pixels across all wavelengths and accelerating the overall detection process.
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 the quantitative measurement of defect sizes smaller than a pixel unit by differentiating between diffuse and specular reflections based on spectral wavelengths, enhancing detection accuracy and precision.
Implementation Method 1
a spectral element that transmits light of a predetermined wavelength and is configured to change the wavelength of the transmitted light
Implementation Method 2
detects a diffuse reflection region by which the illumination light is diffusely reflected
Implementation Method 3
an imaging unit that images the illumination light reflected by the measuring object
Data Source
AI summary
A defect detecting device includes an illumination that irradiates a measuring object with illumination light, an imager that images the illumination light reflected by the measuring object, and a detector that detects a defect at a surface of the measuring object based on a captured image obtained by imaging the illumination light by the imager. The captured image includes a plurality of spectral images having different spectral wavelengths, and the detector detects a diffuse reflection region by which the illumination light is diffusely reflected based on the plurality of spectral images, and determines a size of the defect based on the spectral wavelength of the spectral image in which the diffuse reflection region is detected.