Shape-based image reconstruction using linearized deformations. (1st February 2017)

• Record Type: Journal Article
• Authors: Öktem, Ozan; Chen, Chong; Onur Domaniç, Nevzat; Ravikumar, Pradeep; Bajaj, Chandrajit
• Is Part Of: Inverse problems. Volume 33:Number 3(2017:Mar.)
• Publication Date: 2017-02-01
• Subjects: tomography -- image reconstruction -- indirect image matching -- computational anatomy -- shape analysis -- inverse problems; Inverse problems (Differential equations) -- Periodicals; 515.357
• DOI: 10.1088/1361-6420/aa55af
• Languages: English
• ISSNs: 0266-5611

Abstract

Abstract: We introduce a reconstruction framework that can account for shape related prior information in imaging-related inverse problems. It is a variational scheme that uses a shape functional, whose definition is based on deformable template machinery from computational anatomy. We prove existence and, as a proof of concept, we apply the proposed shape-based reconstruction to 2D tomography with very sparse and/or highly noisy measurements.