Improving the speed of multiresolution segmentation using slic superpixels

Ovidiu Csillik*, Stefan Lang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

134 Downloads (Pure)

Abstract

The popular multiresolution segmentation (MRS) algorithm is time and memory consuming when dealing with large images because it uses the pixel-grid for the initial object representation. In this study, we have tested a new workflow for image segmentation of remote sensing data, starting the MRS (using the ESP2 tool) from the superpixel level (using SLIC superpixels) and aiming at dramatically reducing the amount of time and computational resources needed to automatically partition relatively large datasets of very high resolution (VHR) remote sensing images. Tests were done on Quickbird and WorldView-2 data and the results show that the proposed workflow outperforms the traditional approach (MRS starting from pixels). The computational time was reduced in all cases, the biggest improvement being from 5h 35min to 13 min, for a WorldView-2 scene with 8 bands and an extent of 12.2 million pixels. This also comes with a slight improvement of the geometric accuracy of the extracted objects. This approach has the potential to enhance the automation of big remote sensing data analysis and processing, especially when time is an important constraint.
Original languageEnglish
Title of host publicationProceedings of GEOBIA 2016 : Solutions and synergies, 14-16 September 2016, Enschede, Netherlands
EditorsN. Kerle, M. Gerke, S. Lefevre
Place of PublicationEnschede
PublisherUniversity of Twente, Faculty of Geo-Information Science and Earth Observation (ITC)
Number of pages4
ISBN (Print)978-90-365-4201-2
DOIs
Publication statusPublished - 14 Sept 2016
Externally publishedYes
Event6th International Conference on Geographic Object-Based Image Analysis, GEOBIA 2016: Solutions & Synergies - University of Twente Faculty of Geo-Information and Earth Observation (ITC), Enschede, Netherlands
Duration: 14 Sept 201616 Sept 2016
Conference number: 6
https://www.geobia2016.com/

Conference

Conference6th International Conference on Geographic Object-Based Image Analysis, GEOBIA 2016
Abbreviated titleGEOBIA
Country/TerritoryNetherlands
CityEnschede
Period14/09/1616/09/16
Internet address

Fingerprint

Dive into the research topics of 'Improving the speed of multiresolution segmentation using slic superpixels'. Together they form a unique fingerprint.

Cite this