Multi-scale object-based fuzzy classification for LULC mapping from optical satellite images
- 42 Downloads
In this paper, a multi-scale object-based fuzzy approach is demonstrated for land use/land cover (LULC) classification using high-resolution multi-spectral optical RapidEye and IKONOS images of Lao Cai and Can Tho areas in Vietnam respectively. Optimal threshold for segmentation procedure is selected from rate of change-local variance graph. Object-based fuzzy approach is implemented to identify LULC classes and LULC initial sets, and then the initial sets are classified to final LULC classes. In case of Lao Cai area, normalized difference vegetation index (NDVI), normalized difference water index (NDWI), water index (WI) in object-based are used to generated water, terrace field classes, and built-up and vegetation sets. NDVI, soil index (SI) and red band are used to distinguish built-up set to building, bare land and road classes. NDVI and RedEgde band are inputs to classify rice field and forest classes from vegetation set. In case of Can Tho area, NDWI and WI are generated to water, vegetation, paddy field classes and built-up set, and then built-up set is classified to building, bare land, road, and paddy field classes. The technique is able to create LULC maps of Lao Cai and Can Tho areas with (90.8%, 0.84), and (92.3%, 0.90) classification accuracy and kappa coefficient, correspondingly.
KeywordsFuzzy LULC Local variance Multi-scale segment Object-based GRASS GIS
We are deeply grateful to Dr. Ho Dinh Duan and Dr. Vinayaraj Poliyapram for their comments which are valuable in improving the manuscript. We also thank the anonymous reviewers for their critical and constructive suggestions. The first author would like to express gratitude to Nishimura International Scholarship Foundation (NISF) for award of fellowship to pursue her doctoral research.
This research was mainly prepared and performed by HTD and VR. HTD and VR contributed with ideas and designing the data processing workflow. LXT and GY provided inputs about data processing methodology and field validation of results and revising of the manuscript.
- 4.Schowengerdt, R. A. (2007). Remote sensing: models and methods for image processing (3rd ed.). New York: Elsevier.Google Scholar
- 5.Blaschke, T., Burnett, C., & Pekkarinen, A. (2004). Image segmentation methods for object-based analysis and classification. In S. M. D. Jong & F. D. V. Meer (Eds.), Remote sensing image analysis: including the spatial domain. Remote sensing and digital image processing (Vol. 5). Dordrecht: Springer.Google Scholar
- 13.Wood, T. F., & Foody, G. M. (1993). Using cover-type likelihoods and typicalities in a geographic information system data structure to map gradually changing environments. In R. Haines-Young, D. R. Green & S. H.Cousins (Eds.), Landscape ecology and GIS (pp. 141–146). London: Taylor and Francis.Google Scholar
- 15.Wang, F. (1990). Improve remote sensing imagery analysis through fuzzy information representation. Photogrammetric Engineering and Remote Sensing, 56, 1163–1169.Google Scholar
- 16.Satellite Imaging Corporation. RapidEye satellite sensors. https://www.satimagingcorp.com/satellite-sensors/other-satellite-sensors/rapideye/. Accessed 16 December 2018.
- 19.Satellite Imaging Corporation. IKONOS satellite sensor. https://www.satimagingcorp.com/satellite-sensors/ikonos/. Accessed 16 December 2018.
- 22.Yamagata, Y., Sugita, M., & Yasuoka, Y. (1997). Development of Vegetation-Soil-Water Index algorithms and applications. Journal of the Remote Sensing Society of Japan, 17(1), 54–64.Google Scholar
- 28.Kim, M., Madden, M., & Warner, T. (2008). Estimation of optimal image object size for the segmentation of forest stands with multispectral IKONOS imagery. In T. Blaschke, S. Lang, & G. J. Hay (Eds.), Object-based image analysis—Spatial concepts for knowledge driven remote sensing applications (pp. 291–307). Berlin: Springer.Google Scholar
- 30.Bauer, R. J., & Dahlquist, J. R. (1998). Technical market indicators: Analysis and performance. New York: Wiley.Google Scholar
- 31.Osgeo.org. PyGRASS documentation. https://grass.osgeo.org/grass70/manuals/libpython/pygrass_index.html. Accessed 25 December 2018.
- 35.Neubert, M., Herold, H., & Meinel, G. (2008). Assessing image segmentation quality—Concepts, methods and application. In T. Blaschke, S. Lang, & G. J. Hay (Eds.), Object-based image analysis. Lecture notes in geoinformation and cartography. Berlin: Springer.Google Scholar