A Transport for imaging process
This work addresses the problem of image interpolation for medical imaging, specifically heart tracking, but the method is incremental as it combines existing techniques with a new formulation.
The authors propose a method to generate continuous motion between two segmented heart images by computing an evolution process that transports one image to the other using optical flow, with a fixed-point and space-time least squares formulation. Numerical results in 2D demonstrate efficiency compared to the Dacorogna-Moser transport.
This work originates from a heart's images tracking which is to generate an apparent continuous motion, observable through intensity variation from one starting image to an ending one both supposed segmented. Given two images $ρ_0$ and $ρ_1$, we calculate an evolution process $ρ(t,\cdot)$ which transports $ρ_0$ to $ρ_1$ by using the optical flow. In this paper we propose an algorithm based on a fixed point formulation and a space-time least squares formulation of the transport equation for computing a transport problem. Existence results are given for a transport problem with a minimum divergence for a dual norm or a weighted $H^1_0$-semi norm, for the velocity. The proposed transport is compare with the transport introduced by Dacorogna-Moser. The strategy is implemented in a 2D case and numerical results are presented with a first order Lagrange finite element, showing the efficiency of the proposed strategy.