Ferruh YILDIZ, Mustafa ERDOĞAN, Dijle BOYACI

Pixel- versus object-based classification of forest and agricultural areas from multiresolution satellite images

Managing of natural resources including agriculture and forestry is a very important subject for governments and decision makers. Up-to-date, accurate, and timely geospatial information about natural resources is needed in the management process. Remote sensing technology plays a significant role in the production of this geospatial information. Compared to terrestrial work, the analysis of larger areas with remote sensing techniques can be done on a shorter timescale and at lower cost. Image classification in remote sensing is one of the most popular methods used for the detection of forest and agricultural areas. However, the accuracy of classification changes according to the source and reference data, the classification method, and the producer s knowledge and experience. In this research, the identification of forests and agricultural areas was studied in terms of both their geometry and attribution using different classification methods and different source data. Landsat, Aster, and RapidEye images, which have different spatial and spectral resolution, were used as the source data. Pixel- and object-based classification algorithms were also tested. Classification accuracy results were evaluated at 300 stratified random pixels. It was found that the best overall accuracy was obtained from Aster imagery with object-based classification using the nearest neighbor method. The results also showed that spatial resolution is important for discrimination of classes and spectral resolution is important for definition of features, and confirmed the well-established paradigm of remote sensing that there are no perfect source data or method of classification for all situations

PDF

___

[1] Seder I, Weinkauf R, Neumann T. Knowledge-based databases and intelligent decision support for environmental management in urban systems. Comput Environ Urban 2000; 24: 233-250.
[2] Holland DA, Boyd DS, Marshall P. Updating topographic mapping in Great Britain using imagery from highresolution satellite sensors. ISPRS J Photogramm Remote Sensing 2006; 60: 212-223.
[3] Borak JS, Strahler AH. Feature selection and land cover classification of a MODIS-like dataset for a semi-arid environment. Int J Remote Sens 1999; 20: 919-938.
[4] Chintan AS, Arora MK, Paramod KV. Unsupervised classification of hyperspectral data: an ICA mixture model based approach. Int J Remote Sens 2004; 25: 481-487.
[5] Quattara T, Gwyn QHJ, Dubois JMM. Evaluation of the runoff potential in high relief semi-arid regions using remote sensing data: application to Bolivia. Int J Remote Sens 2004; 25: 423-435.
[6] Günlü A, Ba¸skent EZ, Kadıo˘gulları A˙I, Altun L. Forest site classification using Landsat 7 ETM data: a case study of Ma¸cka-Orman¨ust¨u Forest, Turkey. Environ Monit Assess 2009; 151: 93-104.
[7] Yüksel A, Akay AE, Gundogan R. Using ASTER imagery in land use/cover classification of Eastern Mediterranean landscapes according to CORINE land cover project. Sensors 2010; 8: 1237-1251.
[8] Bouziani M, Goita K, He D. Rule-based classification of a very high resolution image in an urban environment using multispectral segmentation guided by cartographic data. IEEE T Geosci Remote 2010; 48: 3198-3211.
[9] Mathieu R, Aryal J, Chong AK. Object-based classification of Ikonos imagery for mapping large-scale vegetation communities in urban areas. Sensors 2007; 7: 2860-2880.
[10] Casals-Carrasco P, Kubo S, Babu Madhavan, B. Application of spectral mixture analysis for terrain evaluation studies. Int J Remote Sens 2000; 21: 3039-3055.
[11] Blaschke T, Strobl J. Whats wrong with pixels? Some recent development interfacing remote sensing and GIS. GeoBIT/GIS 14 2001; 6: 12-17.
[12] Carleer AP, Wolff E. Region-based classification potential for land-cover classification with very high spatial resolution satellite data. Proceedings of 1st International Conference on Object-based Image Analysis, 45 July 2006; Salzburg University, Austria: pp. 1682-1777.
[13] Blaschke T. Object based image analysis for remote sensing. ISPRS J Photogramm 2010; 65: 2-16.
[14] Shackelford AK, Davis CH. A combined fuzzy pixel-based and object-based approach for classification of highresolution multispectral data over urban areas. IEEE T Geosci Remote 2003; 41: 2354-2364.
[15] Baatz M, Benz U, Dehghani S, Heynen M, H¨oltje A, Hofmann P, Lingenfelder I, Mimler M, Sohlbach M, Weber M et al. e-Cognition Professional User Guide. Munich, Germany: Definiens Imaging GmbH, 2004.
[16] Baatz M, Schape A. Multiresolution segmentation: an optimization approach for high quality multi-scale image segmentation. In: Strobl J, Blaschke T, Griesebner G, editors. Angewandte Geographische Informations- Verarbeitung XII. Karlsruhe, Germany: Wichmann Verlag, 2000; pp. 12-23.
[17] Benz UC, Hofmann P, Willhauck G, Lingenfelder I, Heynen M. Multi-resolution, object oriented fuzzy analysis of remote sensing data for GIS-ready information. ISPRS J Photogram 2004; 58: 239-258.
[18] Hellman M. Fuzzy Logic Introduction. Universit´e de Rennes, France, 2001.
[19] Bardossy A, Samaniego L. Fuzzy rule-based classification of remotely sensed imagery. IEEE T Geosci 2002; 40: 362-374.
[20] Gao J. Digital Analysis of Remotely Sensed Data. New York, NY, USA: McGraw-Hill, 2009.
[21] Oztan NS, S¨uzen ML. Mapping evaporate minerals by ASTER. Int J Remote Sens 2011; 32: 1651-1673. ¨
[22] McFeeters SK. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water areas. Int J Remote Sens 1996; 17: 1425-1432.
[23] Liu L, Zhang Y. Urban heat island analysis using the Landsat TM data and ASTER data: a case study in Hong Kong. Remote Sens 2011; 3: 1535-1552.
[24] Ustuner M, Sanli FB, Abdikan S, Esetlili MT, Kurucu Y. Crop type classification using vegetation indices of RapidEye imagery. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 2014; 7: 195-198.
[25] Yan G, Mas JF, Maathuis BHP, Zhang X, Van Dijk PM. Comparison of pixel-based and object-oriented image classification approaches a case study in a coal fire area, Wuda, Inner Mongolia, China. Int J Remote Sens 2006; 27: 4039-4055.