Radiation Image Harmonic Reduction for Grid Stripe Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for reducing grid stripes in radiation images using grids for scattered radiation reduction suffer from accuracy issues due to high spectral intensity of object components relative to grid components, particularly in higher-order harmonics, leading to deterioration of object components and decreased fitting accuracy.
Innovation Solution
An image processing apparatus that converts the frequency spectrum of a fundamental wave of a grid into an n-th harmonic and reduces the n-th harmonic in the radiation image using an intensity adjustment method, employing conversion and reduction units to estimate and minimize the harmonic without affecting object components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-generated harmful factors
If grid model fitting is applied to reduce grid stripes, then grid stripe reduction is achieved, but accuracy deteriorates when spectral intensity of object component is high relative to grid component
Solution Approach 1:
The invention changes the frequency parameter by converting the fundamental wave frequency spectrum to n-th harmonic frequency spectrum. This parameter transformation allows the system to target and reduce grid stripe components at different frequency levels, improving accuracy when object component spectral intensity is high relative to grid component spectral intensity.
Solution Approach 2:
The invention segments the frequency spectrum into fundamental wave components and n-th harmonic components. By separately processing these segmented frequency components through conversion and selective reduction, the system can accurately target grid stripe patterns without being overwhelmed by object component interference.
2Manufacturing precision
If higher frequency is used for grid fundamental wave, then object component deterioration is prevented, but n-th harmonic frequency becomes low relative to sampling pitch
Solution Approach 1:
The invention addresses the frequency dimension problem by introducing harmonic order as an additional dimension. Instead of only varying fundamental wave frequency, the system converts to n-th harmonic frequency spectra, effectively adding a new dimension (harmonic order n) to the frequency parameter space, allowing simultaneous optimization of object component preservation and grid stripe reduction.
Data Source
AI summary
An image processing apparatus that processes a radiation image captured using a grid for scattered radiation reduction includes a conversion unit configured to convert a frequency spectrum of a signal of a fundamental wave of the grid into a frequency spectrum of an n-th harmonic where n is an integer greater than 1, and a reduction unit configured to acquire a reduction image obtained by reducing the n-th harmonic in the radiation image, using an intensity of the signal obtained by the conversion.