Tomography Rotation Axis Determination Using Iterative Reprojection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current tomography techniques face challenges in accurately determining the rotation axis of objects, leading to suboptimal 3D reconstruction quality due to inaccuracies in rotation movement, especially when the axis is tilted, resulting in arc-shaped artefacts and reliance on operator expertise.

Innovation Solution

A method involving the selection of data subgroups along the presumed rotation axis, repeated reconstruction and reprojection steps to determine translation offsets, and iterative correction to pinpoint the real rotation axis, utilizing robust algorithms like SIRT and cross-correlation for precise determination.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional visual estimation methods are used to determine the rotation axis, then operator expertise can guide the process, but the precision is limited to about one degree and depends on operator skill

Engineering Contradiction:

Improverotation axis determination precisionVSAvoidoperator expertise requirement

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system performs self-alignment by automatically calculating the rotation axis through iterative reconstruction and reprojection comparison, eliminating the need for operator visual estimation and expertise while achieving sub-pixel precision

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual visual estimation process is replaced with an automated computational algorithm that uses iterative reconstruction, reprojection, and cross-correlation calculations to determine the rotation axis position

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If the rotation axis is not perfectly known or is tilted, then 3D reconstruction can still be performed, but arc-shaped artefacts appear on the reconstructed sections

Engineering Contradiction:

Improvehandling of tilted rotation axesVSAvoidarc-shaped artefacts

Core Design Contradiction:

Adaptability or versatilityVSObject-generated harmful factors

Solution Approach 1:

The system uses feedback from comparing reprojections with original projections to iteratively refine the rotation axis position, calculating offsets and adjusting the axis until artefacts are minimized and alignment is optimized

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary alignment by determining the rotation axis and calculating translation offsets before performing the final 3D reconstruction, preventing artefact formation in the reconstructed sections

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If iterative reconstruction and reprojection methods are used to determine the rotation axis, then precision is improved, but the computational work and time increase

Engineering Contradiction:

Improverotation axis determination precisionVSAvoidprocessing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system performs a limited number of iterative reconstruction and reprojection cycles sufficient to achieve sub-pixel precision without excessive computation, balancing accuracy with processing efficiency

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system extracts only the essential information needed for rotation axis determination from the projection data, performing calculations on selected subgroups rather than processing all data points exhaustively

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS9881370B2Method of determining a rotation axis of an object in tomography and characterisation method by tomography

Publication Date: 2018.01.30 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES

AI summary

Method of determining a rotation axis of an object during acquisition of projections obtained by a tomography characterization system, comprising the following steps, starting from said projections:Select zones on the projections along the presumed rotation axis;Reconstruct a structure of the object using data in the selected zones,Make reprojections of the reconstruction;Determine a total offset for each selected zone;Determine the real rotation axis using total offsets determined for all zones.