A Completion Network for Reconstruction from Compressed Acquisition

Nicolas Ducros (1) , A Lorente Mur (1) , F. Peyrin (1)

1 Imagerie Tomographique et Radiothérapie

We consider here the problem of reconstructing an image from a few linear measurements. This problem has many biomedical applications, such as computerized tomography, magnetic resonance imaging and optical microscopy. While this problem has long been solved by compressed sensing methods, these are now outperformed by deep-learning approaches. However, understanding why a given network architecture works well is still an open question. In this study, we proposed to interpret the reconstruction problem as a Bayesian completion problem where the missing measurements are estimated from those acquired. From this point of view, a network emerges that includes a fully connected layer that provides the best linear completion scheme. This network has a lot fewer parameters to learn than direct networks, and it trains more rapidly than image-domain networks that correct pseudo inverse solutions. Although, this study focuses on computational optics, it might provide some insight for inverse problems that have similar formulations.

Mots clés: fully connected layer; Image reconstruction; Deep learning; Bayesian completion; Fully Connected Layer; Index Terms-Image reconstruction; deep learning

Domaines: Imagerie médicale; Réseau de neurones [cs.NE]; Optique / photonique; Traitement du signal et de l'image [eess.SP]

Nicolas Ducros, A Lorente Mur, F. Peyrin. A Completion Network for Reconstruction from Compressed Acquisition. 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), Apr 2020, Iowa City, United States. pp.619-623, ⟨10.1109/ISBI45749.2020.9098390⟩. ⟨hal-02342766v2⟩