Image Processing Apparatus for Periodic Pattern Inclination Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting the inclination of periodic patterns in images, such as those caused by grids in radiography, are either inaccurate due to directional limitations or have high computational loads, particularly when using two-dimensional Fourier transformations.
Innovation Solution
An image processing apparatus and method that employs one-dimensional filtering in multiple directions to remove low-frequency components and detect frequency components corresponding to periodic patterns, allowing for accurate inclination detection with a lower calculation load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If two-dimensional Fourier transformation is used to detect spatial frequency components, then detection accuracy is improved, but calculation load increases
Solution Approach 1:
The patent segments the two-dimensional frequency detection problem into multiple one-dimensional frequency detections along different directions. By dividing the detection into directional components (e.g., horizontal, vertical, and diagonal directions), the method achieves comprehensive frequency detection while reducing computational complexity compared to full two-dimensional Fourier transformation.
Solution Approach 2:
The patent transforms the two-dimensional detection problem into a series of one-dimensional detection problems by introducing directional dimensions. Instead of performing a single comprehensive two-dimensional analysis, the method performs multiple one-dimensional analyses along different angular directions, effectively reducing the dimensionality of each individual calculation while maintaining overall detection accuracy.
2Device complexity
If filtering is performed only in horizontal and vertical directions, then device complexity is reduced, but measurement precision of periodic pattern inclination deteriorates
Solution Approach 1:
The patent introduces dynamic adaptability by allowing the selection of filtering directions based on the actual orientation of the periodic pattern. Instead of being limited to fixed horizontal and vertical directions, the method can dynamically adjust the filtering directions to match the pattern's orientation, improving inclination detection accuracy while maintaining relatively simple device complexity.
Solution Approach 2:
The patent creates a universal filtering system that can handle periodic patterns in any orientation by implementing multi-directional filtering capabilities. The same filtering apparatus can be applied horizontally, vertically, and diagonally, making the device universally applicable to different pattern orientations without requiring separate specialized systems for each direction.
Data Source
AI summary
A subject component removed signal is obtained from an image signal representing an image by performing, with respect to a first direction and/or a second direction that is different from the first direction, one-dimensional filtering processing using a subject component removal filter that removes a low frequency component including a subject component of the image to roughly remove the subject component. Further, frequency components corresponding to the periodic pattern in the subject component removed signal are detected with respect to both of the first direction and the second direction.