Image Filtering with Variable Tap Spacing for Frequency Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional spatial filtering techniques face challenges in controlling high frequencies while reducing the number of reference pixels, leading to decreased high frequency controllability and increased calculation amounts due to larger filter sizes.
Innovation Solution
The technique employs multiple filters with different filter sizes and arrangement spacings for significant coefficients, allowing for control from low to high frequencies by sequentially applying filters with decreasing tap numbers and arrangement spacings, thereby maintaining efficient calculation processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the size of the reference area (filter size) is increased to improve filtering characteristics, then the filtering performance is improved, but the calculation amount increases
Solution Approach 1:
The patent segments the filtering process into multiple stages, applying different filters with different reference area sizes to different frequency components. Low-pass filters with large reference areas handle low frequencies, while high-pass filters with small reference areas handle high frequencies, avoiding the need to apply a single large filter to the entire image
Solution Approach 2:
The patent applies different filter characteristics to different spatial frequencies locally. By separating low-frequency and high-frequency components and applying appropriate filters to each, the system optimizes filtering performance for each frequency band while minimizing overall calculation requirements
2Productivity
If the tap number is reduced to decrease the calculation amount, then the processing speed is improved, but the high frequency controllability decreases
Solution Approach 1:
The patent segments frequency control into separate low-pass and high-pass filtering operations. Each filter is designed with an appropriate tap number for its specific frequency range, allowing the high-pass filter to maintain high frequency controllability with a smaller tap number while the low-pass filter handles low frequencies with a larger tap number
Solution Approach 2:
The patent changes the parameters (filter type, reference area size, tap number) based on the frequency component being processed. By adapting filter parameters to the specific frequency band, the system maintains optimal controllability for each frequency range while minimizing overall calculation requirements
3Productivity
If the arrangement spacing of significant coefficients is increased to reduce the number of reference pixels, then the calculation amount is reduced, but the high frequency controllability decreases
Solution Approach 1:
The patent segments the frequency spectrum into low-frequency and high-frequency bands, applying different filtering strategies to each. This segmentation allows sparse filtering (increased arrangement spacing) to be applied to low-frequency components where it is effective, while dedicated high-pass filters with appropriate spacing handle high-frequency components
Solution Approach 2:
The patent applies different coefficient arrangement spacing to different frequency components. Low-frequency filtering uses sparse arrangements that reduce calculation amount, while high-frequency filtering uses arrangements optimized for edge detection and detail preservation, ensuring high frequency controllability is maintained where needed
Data Source
AI summary
An image that is to be a target for filtering processing is acquired, a plurality of filters having mutually different filter sizes and arrangement spacing for significant coefficients in the filters are applied to the image to perform filtering processing on the image.