Strip Surface Defect Detection With Region-Based Shading Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional shading correction methods for detecting surface defects in steel materials struggle with accuracy due to issues like 'recoil' around defective areas, 'hollowing out' of large defects, and inadequate correction near edges, especially when the material edges shift or width changes.
Innovation Solution
A surface defect detection method and device that calculates an average image from a series of images, recognizes the inspection target region, and performs shading correction only within this region, using either division or subtraction methods to maintain uniform luminance levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If conventional shading correction methods are applied to the entire image, then luminance uniformity is improved, but defect detection accuracy deteriorates due to recoil and hollowing out effects
Solution Approach 1:
The patent divides the image into multiple regions (inspection target region and non-inspection region) and applies shading correction only to the inspection target region. This segmentation prevents the correction process from affecting non-defect areas, thereby avoiding recoil and hollowing out effects while maintaining luminance uniformity in the regions that require it.
Solution Approach 2:
The patent applies different processing treatments to different regions of the image. The inspection target region receives shading correction to achieve luminance uniformity, while the non-inspection region is excluded from correction to preserve original defect characteristics. This local differentiation resolves the contradiction between uniformity and accuracy.
2Ease of operation
If shading correction is applied using fixed patterns, then processing simplicity is improved, but correction accuracy deteriorates when material edges shift or width changes
Solution Approach 1:
The patent dynamically determines the inspection target region for each image based on the actual material position and dimensions in that specific image. This dynamic approach allows the shading correction to adapt to edge shifts and width changes, maintaining correction accuracy without requiring complex pre-calibration of fixed patterns.
Solution Approach 2:
The patent performs preliminary identification of the inspection target region in each image before applying shading correction. By pre-defining the correction region based on actual material boundaries, the system ensures that subsequent correction operations are both simple to execute and accurate to the material's actual state.
3Measurement precision
If the inspection target region is dynamically recognized, then defect detection accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent segments the image processing into distinct steps: identifying the inspection target region, then applying shading correction only to that region. This segmentation simplifies the overall process by clearly defining where correction should and should not be applied, reducing unnecessary processing while maintaining accuracy.
Solution Approach 2:
The patent applies local processing by restricting shading correction to only the inspection target region rather than the entire image. This local approach improves accuracy by focusing computational resources where needed while reducing overall processing complexity by excluding non-relevant areas from correction operations.
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
Accurately detects surface defects by suppressing 'recoil' and 'hollowing out', ensuring precise defect classification even with edge shifts or width variations.
Implementation Method 1
detecting reflected light from the strip-shaped body obtained by illuminating a surface of the strip-shaped body
Data Source
AI summary
A surface defect detection method for optically detecting a surface defect in a strip-shaped body includes an image acquisition step of detecting reflected light from the strip-shaped body obtained by illuminating a surface of the strip-shaped body and imaging while relatively scanning the surface of the strip-shaped body to acquire a plurality of images including the surface of the strip-shaped body, an average image calculation step of calculating an average image of the acquired images, an image correction step of performing shading correction on each acquired image using the average image to obtain corrected images, and a defect detection step of detecting a surface defect in the strip-shaped body based on the corrected images. The average image calculation step includes recognizing an inspection target region in which the strip-shaped body is located in each image and contributing to the average image only for pixels in the inspection target region.