Trainable Regularization for Multi-frame Superresolution

Teresa Klatzer, Daniel Soukup, Erich Kobler, Kerstin Hammernik, Thomas Pock

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present a novel method for multi-frame superresolution (SR). Our main goal is to improve the spatial resolution of a multi-line scan camera for an industrial inspection task. High resolution output images are reconstructed using our proposed SR algorithm for multi-channel data, which is based on the trainable reaction-diffusion model. As this is a supervised learning approach, we simulate ground truth data for a real imaging scenario. We show that learning a regularizer for the SR problem improves the reconstruction results compared to
an iterative reconstruction algorithm using TV or TGV regularization. We test the learned regularizer, trained on simulated data, on images acquired with the real camera setup and achieve excellent results.
Original languageEnglish
Title of host publicationPattern Recognition
Subtitle of host publicationGerman Conference, GCPR 2017, Proceedings
EditorsV. Roth, T. Vetter
PublisherSpringer
Pages90-100
ISBN (Print)978-3-319-66708-9
DOIs
Publication statusPublished - 2017

Publication series

NameLecture Notes in Computer Science
Volume10496

Fingerprint Dive into the research topics of 'Trainable Regularization for Multi-frame Superresolution'. Together they form a unique fingerprint.

  • Cite this

    Klatzer, T., Soukup, D., Kobler, E., Hammernik, K., & Pock, T. (2017). Trainable Regularization for Multi-frame Superresolution. In V. Roth, & T. Vetter (Eds.), Pattern Recognition: German Conference, GCPR 2017, Proceedings (pp. 90-100). (Lecture Notes in Computer Science; Vol. 10496). Springer. https://doi.org/10.1007/978-3-319-66709-6_8