Printed Character Defect Detection Using Skeleton Contours
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect detection technologies struggle to accurately identify shape defects in complex printed characters like Chinese and Japanese characters, leading to over-correction issues due to inadequate characterization of their shape features.
Innovation Solution
A method involving image processing steps including binarization, thinning, and contour analysis using OpenCV algorithms to compare the number of contours in reference and detected character images, ensuring accurate detection of defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing defect detection technology is used for complex printed characters with many strokes, then detection can be performed, but over-correction occurs due to inability to accurately describe shape features
Solution Approach 1:
The patent transforms the character image into a skeleton structure through thinning operations, changing the parameter representation from pixel-based images to topological skeleton graphs. This parameter transformation enables accurate shape feature extraction while reducing the complexity of preprocessing procedures needed for defect detection in complex characters
Solution Approach 2:
The patent segments the character detection process into distinct phases: image binarization, skeleton thinning, and contour analysis. By segmenting the complex detection task into manageable steps with clear intermediate representations, the system achieves accurate shape feature description without requiring overly complex preprocessing procedures
2Measurement precision
If Hu moments are used for defect detection, then detection can be performed, but shape features of characters cannot be easily described requiring adjustments to preprocessing procedures
Solution Approach 1:
The patent replaces the Hu moments mathematical transformation approach with a skeleton-based topological analysis method. This substitution provides more intuitive and accurate shape feature description through contour analysis of thinned character structures, eliminating the need for preprocessing adjustments required by Hu moments
3Measurement precision
If existing defect detection technology is used for complex characters, then detection can be performed, but Hu moments do not provide adjustment directions
Solution Approach 1:
The patent implements a feedback mechanism where the skeleton-based contour analysis provides clear adjustment directions by comparing the number and characteristics of contours between reference and detected characters. This feedback loop enables operators to understand and correct detection issues systematically, unlike Hu moments which lack explicit adjustment guidance
Data Source
AI summary
This application provides a method of detecting printing defects. The method includes obtaining a first image of each character in a reference image. A third image of each character is obtained based on the first image of each character, a fourth image of each character is obtained based on a second image of each character obtained from an image to be detected. Once a fifth image of each character is obtained based on the third image of each character, a sixth image of each character is obtained according to the fourth image and the fifth image of each character, a detection result of each character in the image to be detected is determined according to the fifth image and the sixth image of the each character.