Improving deep learning-based segmentation of diatoms in gigapixel-sized virtual slides by object-based tile positioning and object integrity constraint

• Michael Kloster ,

  Roles Conceptualization, Data curation, Investigation, Methodology, Software, Validation, Writing – original draft

  * E-mail: michael.kloster@uni-due.de

  Affiliation Department of Phycology, Faculty of Biology, University of Duisburg-Essen, Essen, Germany

  https://orcid.org/0000-0001-9244-4925

  ⨯
• Andrea M. Burfeid-Castellanos,

  Roles Data curation, Writing – original draft

  Affiliation Department of Phycology, Faculty of Biology, University of Duisburg-Essen, Essen, Germany

  https://orcid.org/0000-0002-8556-3157

  ⨯
• Daniel Langenkämper,

  Roles Conceptualization, Writing – original draft

  Affiliation Biodata Mining Group, Faculty of Technology, Bielefeld University, Bielefeld, Germany

  ⨯
• Tim W. Nattkemper,

  Roles Conceptualization, Funding acquisition, Supervision, Writing – original draft

  Affiliation Biodata Mining Group, Faculty of Technology, Bielefeld University, Bielefeld, Germany

  https://orcid.org/0000-0002-7986-1158

  ⨯
• Bánk Beszteri

  Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – original draft

  Affiliation Department of Phycology, Faculty of Biology, University of Duisburg-Essen, Essen, Germany

  https://orcid.org/0000-0002-6852-1588

  ⨯