Find Rooms for Improvement: Towards Semi-automatic Labeling of Occupancy Grid Maps

Sven Hellbach, Marian Himstedt, Frank Bahrmann, Martin Riedel, Thomas Villmann, Hans-Joachim Böhme


Semi-automatic semantic labeling of occupancy grid maps has numerous applications for assistance robotic. This paper proposes an approach based on non-negative matrix factorization (NMF) to extract environment specific features from a given occupancy grid map. NMF also computes a description about where on the map these features need to be applied. We use this description after certain pre-processing steps as an input for generalized learning vector quantization (GLVQ) to achieve the classification or labeling of the grid cells. For the supervised training of the GLVQ the assigned label is propagated to all grid cells of a semantic unit using a simple, yet effective segmentation algorithm. Our approach is evaluated on a standard data set from University of Freiburg, showing very promising results.
Original languageEnglish
Title of host publicationNeural Information Processing
EditorsChu Kiong Loo, Keem Siah Yap, Kok Wai Wong, Andrew Teoh Beng Jin, Kaizhu Huang
Number of pages10
Place of PublicationCham
PublisherSpringer International Publishing
Publication date01.11.2014
ISBN (Print)978-3-319-12642-5
ISBN (Electronic)978-3-319-12643-2
Publication statusPublished - 01.11.2014
Externally publishedYes
Event21st International Conference on Neural Information Processing - Kuching, Malaysia
Duration: 03.11.201406.11.2014


Dive into the research topics of 'Find Rooms for Improvement: Towards Semi-automatic Labeling of Occupancy Grid Maps'. Together they form a unique fingerprint.

Cite this