Artifact-Reduced Voxel Data via Orthogonal CT Scans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computed tomography (CT) methods suffer from artefacts, particularly reconstruction and beam hardening artefacts, which degrade the quality of data records and complicate the identification of structures, especially in industrial CT scans of metallic objects, and previous methods are either time-consuming or limited to correcting non-metallic parts.
Innovation Solution
A method involving the acquisition of CT images from multiple angles with the object tilted orthogonally between scans, using an iterative image data reconstruction algorithm that combines these data sets to generate an artefact-reduced voxel data record, employing techniques like maximum likelihood expectation maximization (MLEM) and data fusion to align and reduce artefacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If previous methods are used to reduce artefacts, then artefact reduction is achieved, but the process is very time and computationally intensive
Solution Approach 1:
The patent introduces a temporal dimension by acquiring CT images at multiple time points, where the object is in different positions due to movement between acquisitions. This temporal-m spatial data is then integrated through image fusion to reduce artefacts, transforming a 3D spatial problem into a 4D spatiotemporal solution that reduces processing time while maintaining artefact reduction quality
2Object-affected harmful factors
If previous methods are used to reduce artefacts, then some artefacts are corrected, but only non-metallic parts can be corrected
Solution Approach 1:
The patent creates a universal artefact reduction method that works for both metallic and non-metallic objects by using image fusion of multiple CT acquisitions. The method does not require material-specific corrections but rather uses the combined information from multiple scans to reduce all types of artefacts, making it applicable to diverse materials including metals, plastics, and composites
3Object-affected harmful factors
If multiple CT images are acquired from different angles, then artefact reduction is improved, but the acquisition time and data processing complexity increase
Solution Approach 1:
The patent uses image fusion techniques that treat multiple CT acquisitions as complementary copies of the same object from different time points and angles. By fusing these copies through algorithmic integration rather than complex manual processing, the method reduces artefacts while keeping data processing manageable through automated fusion algorithms
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly reduces artefacts by aligning data from orthogonal orientations, improving the accuracy and quality of CT data records with fewer interpolation steps, thus enhancing the detection of structures and reducing computational intensity.
Implementation Method 1
X-ray computed tomography (CT) is a method for obtaining information about the interior of objects
Implementation Method 2
In x-ray computed tomography, artefacts arise as a result of various effects, e.g. as a result of the employed reconstruction method and as a result of beam hardening
Data Source
AI summary
The present invention relates to a method and a computer program product for generating an artifact-reduced voxel data record of an object. The artifact-reduced voxel data record of the object is generated with the aid of a computed tomography scanner. In some aspects of the present disclosure, a first image data record and a second image data record is generated by acquiring computed tomography images of the object. In still other aspects of the present disclosure, the artifact-reduced voxel data record of the object is generated with the aid of an image data reconstruction algorithm.