Enhanced evaluation of image segmentation results

Abstract

Geographic object-based image analysis (GEOBIA) is a promising methodology for image analysis, in which images are first segmented into image segments (or objects) and then analysed based on shape, texture, context and spectral features. The extra dimension of data offered by the objects yields a more enhanced image analysis. The first and most important step is thus the segmentation of images. The effectiveness of the object-based image analysis depends entirely on the quality of the segmentation result. There exist several types of image segmentation algorithms developed for a variety of applications ranging from medical imaging to remote sensing image analysis. It is, therefore, necessary to have an evaluation measure to decide which algorithm can be better for a particular task. Like segmentation itself, there is no standard way of evaluating segmentation results. In this article, we provide an easy way to analyse segmentation results by defining what qualifies as under-segmentation and over-segmentation while analysing the segmentation results of user-selected reference regions. The evaluation criteria are designed to handle the results of multi-level segmentation algorithms, which are commonly used in GEOBIA.

Keywords:

• remote sensing
• image segmentation
• object-based image analysis
• segmentation evaluation
• parameter optimisation
• object delineation quality

Acknowledgements

We would like to express our gratitude to Chris Oliver (InfoSAR Limited, UK), Carsten Steger (Munich University of Technology/MVTec GmbH, Germany), James C. Tilton (Goddard Space Flight Center, National Aeronautics and Space Administration, USA), Peijun Li (Peking University, Institute of Remote Sensing and GIS, China), Guillermo Castilla (University of Calgary, Department of Geography, Canada), Arko Lucieer (University of Tasmania, Centre for Spatial Information Science, Australia), Manoel de Araújo Sousa Jr. (Instituto National de Pesquisas Espacias, Brazil) and Emanuel Gofman (IBM Haifa Research Labs, Israel), for their collaboration, their technical support for the segmentation algorithms and their assistance in processing our sample data. Furthermore, we would like to thank ITT Visual Information Solutions (Boulder, USA) as well as CREASO (Gilching, Germany) for making available a test version of the ENVI Feature Extraction Module. We also would like to express our gratitude to the unknown reviewers as well as to the editor of this special issue, Kasper Johansen, who helped us to improve the quality of this article.