X-ray CT Image Processing for Streak Artifact Removal and Edge Preservation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques for X-ray CT images struggle to maintain the edge of structures while effectively removing streak artifacts, as they often incorrectly classify streak artifacts as edges or non-flat regions, leading to blurred structures in clinical diagnostics.
Innovation Solution
An image processing device and method that uses a weighting coefficient calculated from a nonlinear function based on feature amounts of original and smoothed images to perform weighted addition, ensuring the edge of structures is maintained and streak artifacts are removed, by determining the shape of the nonlinear function and calculating the weighting coefficient to meet specific constraint conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If projection data smoothing filter is applied to reduce streak artifacts, then streak artifact reduction effect is improved, but structure edge sharpness deteriorates
Solution Approach 1:
The patent applies different processing strategies to different regions of the image. Flat regions undergo smoothing to remove streak artifacts, while non-flat regions (containing structure edges) are preserved without smoothing. This is achieved by calculating a determination value for each pixel that identifies whether it belongs to a flat or non-flat region, enabling localized quality optimization.
Solution Approach 2:
The image is segmented into flat regions and non-flat regions based on the determination value calculated from pixel data. This segmentation allows the system to apply appropriate processing to each region type, separating the streak artifact removal function from the edge preservation function.
2Manufacturing precision
If edge detection is performed using pixel value differences and variance to perform weighted addition, then structure edge maintenance is improved, but streak artifacts are erroneously detected as edges leading to incorrect classification
Solution Approach 1:
The patent changes the parameter used for region determination from traditional edge detection metrics (pixel value differences, variance) to a new determination value that is insensitive to streak artifacts. This parameter change enables accurate discrimination between actual structure edges and artifact patterns.
Solution Approach 2:
The patent converts the characteristic of streak artifacts (linear structure-like appearance) from a harmful factor that causes misclassification into a distinguishable feature. By designing the determination value calculation to recognize the specific pattern of streak artifacts, the system can identify and exclude them from edge detection, turning their visual similarity to edges into a detectable signature.
Data Source
AI summary
In order to provide an image processing device and the like making it possible to generate a target image in which edges of a structure are upheld and from which streaking artifacts are removed, a computation device determines a shape of a non-linear function on the basis of feature amounts of an original image and a smoothed image (S101). Next, the computation device calculates a condition coefficient of the original image and the smoothed image by using the non-linear function for which the shape was determined in S101 (S102). Next, the computation device uses the condition coefficients calculated in S102 to calculate a weighting coefficient for each of the pixels of the original image and the smoothed image (S103). Next, the computation device adds weighting to the original image and the smoothed image to generate the target image (S104).