Printed Character Misprint Detection Using Fourier Descriptors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Defects in printed characters, such as misprints, affect product quality and can mislead users, necessitating effective detection methods.
Innovation Solution
A method using a computing device to analyze reference and test images of characters, involving image processing, optical character recognition, and Fourier descriptors to determine similarity between outline images, with thresholds for defect detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image comparison methods are used to detect misprinted characters, then the detection process is simple, but the detection precision is low and cannot accurately identify subtle defects
Solution Approach 1:
The patent segments the character image into multiple regions (e.g., radical regions, component regions) and performs separate image processing and comparison on each region. This segmentation allows the system to detect subtle defects in specific areas without being overwhelmed by the entire character image, thereby improving detection precision while managing complexity through localized analysis.
Solution Approach 2:
The patent transforms the character images into Fourier domain representations using Fourier descriptors, converting spatial domain image data into frequency domain features. This dimensional transformation enables the system to capture subtle structural differences and defects that are not apparent in the original image space, significantly improving detection precision for misprinted characters.
2Measurement precision
If detailed image processing is performed to improve defect detection accuracy, then the detection precision increases, but the processing time increases
Solution Approach 1:
The patent extracts key features from the character images using Fourier descriptors, which capture the essential structural information while discarding redundant pixel-level details. This extraction process reduces the data volume that needs to be processed and compared, thereby improving defect detection accuracy through focused feature analysis while reducing overall processing time.
Solution Approach 2:
By segmenting the character image into meaningful regions and applying image processing only to these segments rather than the entire image, the system achieves high detection accuracy in critical areas while minimizing the total computational workload and processing time required.
3Reliability
If the detection system uses simple comparison methods, then the processing speed is fast, but the reliability of defect detection is low
Solution Approach 1:
The transformation to Fourier domain provides a more reliable basis for comparison by representing character structures in terms of their frequency components. This dimensional change enables the system to detect subtle defects with high reliability while maintaining processing efficiency through mathematical transformations that are computationally tractable.
Solution Approach 2:
The patent changes the parameter representation from pixel values to Fourier descriptors, which are more sensitive to structural variations and defects. This parameter transformation improves detection reliability by capturing essential character features in a form that is more suitable for comparison and defect identification, while the efficient computation of Fourier descriptors maintains processing speed.
Data Source
AI summary
A method of detecting misprints obtains a reference image of perfectly-formed characters and a test image showing the same characters. First image processing is performed on the reference image to obtain a first image, and first image processing is performed on the test image to obtain a second image. A second image processing is performed on the first image to obtain a first outline image of each of first characters in the reference image, and the second image processing is performed on the second image to obtain a second outline image of each of second characters in the test image. A corresponding first outline image is determined for the second outline image. A similarity between the corresponding first outline image and the second outline image is calculated. Accordingly, a detection result of the second outline image is determined. The method can detect character flaw accurately.