Skip to Content

Resource hub

Review of image segmentation algorithms for analysing Sentinel-2 data over large geographical areas 2020

Abstract

Earth Observation (EO) has been extensively used to provide a synoptic view of land use, cover and change at a variety of scales. As methods to analyse imagery evolve an important aspect is a reliable spatial framework, referred to as a segmentation, to act as the basis of a classification. A segmentation is part of a method referred to as object-based image analysis (OBIA), where spatially and spectrally homogenous units are identified and then grouped, classifying areas of similar properties. JNCC, on behalf of Natural England, carried out an investigation into methods of deriving OBIA from open-source and proprietary packages currently available.

This report outlines the process of deriving OBIA from open-source and proprietary packages and provides recommendations for deployment. Implications and future research are provided.

Resource type Publication

Topic category Environment

Reference date 2020·07·27

Citation
Sideris, K., Colson, D., Lightfoot, P., Heeley, L. & Robinson, P. 2020. Review of image segmentation algorithms for analysing Sentinel-2 data over large geographical areas. JNCC Report No. 655, JNCC, Peterborough, ISSN 0963-8091.

Lineage
This research was carried out with financial support from Natural England. We would like to thank our colleagues at JNCC for their support and input into the project. We would also like to thank the Rural and Environmental Science & Analytical Services (RESAS) team in the Scottish Government for providing access to data sources. Finally, we would like to extend our thanks and acknowledge the Evidence Earth Observation Service (EEOS) in Natural England for their support throughout the project.

Responsible organisation
Communications, JNCC publisher

Limitations on public access No limitations

Use constraints Available under the Open Government Licence 3.0

Metadata date 2020·07·28

Metadata point of contact
Communications, JNCC

Back to top