Object-oriented image analysis to extract landscape elements in urban fringes, Central Japan
- 375 Downloads
A method enabling the object-oriented image analysis of landscape elements incorporating topographic data was designed and tested on a Japanese countryside target area. IKONOS data (four multispectral bands with a spatial resolution of 4 m and a panchromatic band with a spatial resolution of 1 m) acquired on 23 April 2001 were used. Definiens v.5 software (Definiens AG, München, Germany) was employed for the classification. The initial segmentation was multiresolution and bottom-up, and each segment identified was considered to be one object. Two classifications employing the same landscape elements and ground truth data were implemented. One classification adopted an object-based image analysis classification method based on spectral characteristics; the other utilized an object-oriented image analysis (OOIA) that allows for a suitable scale parameter to be selected independently for each landscape element. In addition, topographic data derived from field surveys (walking surveys) and topographic maps were used to create a topographic database delineating the boundary between valley bottoms and the adjacent slopes (elevation: about 10 m). These data were then integrated into the OOIA analysis. The accuracies of the two classifications were assessed by comparing the results to a master landscape map produced directly from aerial photographs and on-site observations. The object-oriented method using the topographic data resulted in a higher overall kappa coefficients (0.63–0.47) than the object-based method. These results indicate that object-oriented image analysis of very high resolution data used in combination with topographic data can be an effective tool for landscape classification in Japan, where historical land-use patterns have resulted in finely dissected landscapes.
KeywordsLandscape classification Topographic data Japanese countryside IKONOS VHR satellite data
The authors would like to thank Professor Kevin M. Short of TUIS for helping to improve the English used in this work. This research was funded in part by the MEXT-Supported Program for the Strategic Research Foundation at Private Universities from 2008-2012 (S0801024).
- Allen TFH, Starr TB (1982) Hierarchy: perspectives for ecological complexity. University of Chicago Press, ChicagoGoogle Scholar
- Baatz M, Schäpe A (2000) Multiresolution segmentation: an optimization approach for high quality multi-scale image segmentation. In: Strobl J, Blaschke T (eds) Angewandte Geographische Informations Verarbeitung XII. Wichmann-Verlag, Heidelberg, pp 12–23Google Scholar
- Baatz M, Hoffmann C, Willhauck G (2008) Progressing from object-based to object-oriented image analysis. Object-based image analysis spatial concepts for knowledge-driven remote sensing applications. Springer, New York, pp 29–42Google Scholar
- Definiens AG (2006) Definiens Professional 5 user guide. Definiens AG, MunichGoogle Scholar
- Hara K, Kamagata N, Ishitsuka T, Tomita M (2007) Object-based classification of rural landscapes using remotely sensed data of various resolutions. In: Proc 7th IALE World Congr, Wageningen, The Netherlands, 8–12 July 2012, Pt 2, p 642Google Scholar
- Kamagata N, Hara K, Mori M, Akamatsu Y, Li Y, Hoshino Y (2008) Object-based classification of IKONOS data for vegetation mapping in Central Japan. Object-based image analysis spatial concepts for knowledge-driven remote sensing applications. Springer, New York, pp 459–476Google Scholar
- O’Neill RV, DeAngelis DL, Waide JB, Allen TFH (1986) A hierarchical concept of ecosystems. Princeton University Press, PrincetonGoogle Scholar
- Wu J (1999) Hierarchy and scaling: extrapolating information along a scaling ladder. Can J Remote Sens 25:367–380Google Scholar
- Yu Q, Gong P, Clinton N, Biging G, Kelly M, Schirokauer D (2006) Object-based detailed vegetation classification with airborne high spatial resolution remote sensing imagery. Photogramm Eng Remote Sens 72:799–811Google Scholar