Tomographic reconstruction of incompressible flow

Area of Science:

• Fluid dynamics
• Image reconstruction
• Scientific computing

Background:

• Tomography reconstruction problems are often under-determined, with multiple solutions fitting the data.
• Fast sampling rates in dynamic systems allow for the incorporation of physical laws as constraints.

Purpose of the Study:

• To demonstrate that enforcing incompressible flow improves the fidelity of tomography reconstructions.
• To show that density moment conservation ensures incompressibility and enhances accuracy.

Main Methods:

• Incorporating incompressible flow as a constraint in the reconstruction process.
• Ensuring incompressibility by enforcing the conservation of density moments.

Main Results:

• The "incompressible" reconstruction method significantly improves accuracy compared to standard methods.
• Conservation of density moments leads to the conservation of the density histogram throughout the reconstructed sequence.

Conclusions:

• Enforcing incompressible flow is a powerful technique for improving dynamic tomography reconstructions.
• Density moment conservation offers a robust way to ensure incompressibility and enhance reconstruction fidelity.