Appearance Inspection Brightness Correction for Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The brightness of inspection images for products can vary due to changes in surface color or lighting conditions, affecting the detection accuracy of predetermined algorithms in appearance inspection methods.
Innovation Solution
An appearance inspection method that adjusts the brightness of the inspection area to a predetermined value by calculating an average brightness within a region of interest, applying a correction gain value to each pixel, and separately correcting brightness for each color channel of the image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If brightness correction is not applied to inspection images, then the inspection process is simple and fast, but the detection accuracy deteriorates due to brightness variations from lighting changes or surface color changes
Solution Approach 1:
The patent applies brightness correction as a preliminary processing step before defect detection. By calculating the average brightness of the inspection area and applying a correction gain to normalize brightness to a target value, the system prepares the image in advance for accurate detection, eliminating the need for complex adaptive detection algorithms
Solution Approach 2:
The patent changes the brightness parameter of the inspection image by applying a correction gain value calculated from the ratio of target brightness to average brightness. This parameter adjustment compensates for lighting variations and surface color changes, maintaining consistent detection conditions without adding complex hardware
2Measurement precision
If brightness correction is applied to the entire image, then detection accuracy is maintained, but processing time and computational load increase
Solution Approach 1:
The patent applies brightness correction selectively only to the region of interest (inspection area) rather than the entire image. By extracting the ROI and applying correction only within this bounded area, the system maintains detection accuracy where needed while reducing computational load and processing time for irrelevant areas
Solution Approach 2:
The patent segments the image into the region of interest and other areas, applying brightness correction only to the ROI. This segmentation allows the system to focus computational resources on the critical inspection area, balancing accuracy requirements with processing efficiency
Data Source
AI summary
An appearance inspection method of a product according to embodiments of the present invention may include obtaining and pre-processing an image of appearance of a target object; pre-processing the image; extracting a region of interest from the preprocessed image and obtaining a brightness-corrected image by performing brightness correction on the extracted region of interest; extracting defective candidates from the brightness-corrected image; obtaining a post-processed image by post-processing the extracted defective candidates; and determining whether the target object is defective based on the post-processed image.