Comparison of three visual, object-based, and supervised classification methods of land use/cover mapping in Mollah-Ahmad Watershed, Ardabil

Subject Areas : KARBARI ARAZI

Azad Kakehmami ¹ , Ardavan Ghorbani ²

1 - Graduated MSc of Range Management, University of Mohaghegh Ardabili
2 - Associate Professor at the Department of Range & Watershed Management, University of Mohaghegh Ardabili

Received: 2017-07-01 Accepted : 2018-01-06 Published : 2018-02-20

Keywords: Supervised classification, Visual interpretation, OLI Sensor, Object-based Classification, QuickBird Satellite,

Abstract :

Due to the growing population and consequently the degradation and uncontrolled use of land, awareness of the state of the land and its use is necessary. In this study, the main purpose was land use /cover mapping of the Mollah Ahmad Watershed. Two types of Google Earth images (Quickbird images, 2013) and Operational Land Imager (OLI) sensor of Landsat 8 image (2014) with three methods of visual interpretation (Google Earth images), object-based classification (Google Earth images), and supervised classification (Landsat 8) were used. In order to evaluate the land use /cover map produced from the methods, 51 points were selected as control points for the accuracy assessment. The results showed that the overall accuracy of the map generated from visual interpretation, object-based classification, and supervised classification were 100, 90, and 72 percent, respectively and their kappa coefficients were 1, 0.85 and 0.6, respectively, which the high accuracy of the maps generated from two methods of visual and object-based interpretation. The method of interpreting the visually using the high-resolution images (Quickbird with 0.65 to 2.9 m resolutions) of Google Earth is the most accurate method; however, the object-based classification method due to its low cost in terms of time in large environment has relatively acceptable accuracy.

References:

منابع

1. Ahani, H., A.Ghorbani, M. Rastgar Moghadam, S. Falah Shamsi, & M, Baghernejad, 2009. Land use change detection using satellite imagery Tangeh-Sorkh watershed of Shiraz. Iranian Journal of Natural Resources and Agricultural.16:242-252.


2. Arzani, H., K.h. Mirakhorlou, & S.Z. Hosseini, 2009. Land use mapping using Landsat7 ETM data (Case study in middle catchment’s of Taleghan). Iranian Journal of Range and Desert Research. 16(2):150-160.


3. Aslami, F., A. Ghorbani, B. Sobhani, & M. Panahandeh, 2015. Comparing artificial neural network, support vector machine and object-based methods in preparation land use/cover maps using Landsat-8 images. Journal of Remote Sensing & GIS Techniques in Natural Resources. 6(3): 1-14.


4. Blaschke, T. 2009. Object based image analysis for remote sensing. Journal of Photogrammetry and Remote Sensing, 65(1): 10-21.


5. Butler, D., 2006. Nature Publication: England. The web-wide world. 439: 776-779.


6. Dellepiane, S.G., & P.C. Smith. 1999. Quality assessment of image classification algorithms for land cover mapping: A review and a proposal for a cost based approach. Journal of Remote Sensing. 20: 1461-1486.


7. ERDAS Field Guide, 2014.


8. Foody, G.M., 1992. On the compensation for chance agreement in image classification accuracy assessment. Journal of Photogrammetric Engineering and Remote Sensing. 58: 1459-1460.


9. Garson, G. D., 1998. Neural networks: An introductory guide for social scientists. London: Sage Publications, 208pp.

10. Ghorbani, A., & M. Pakravan, 2013. Land use mapping using visual vs. digital image interpretation of TM and Google earth derived imagery in Shrivan-Darasi watershed (Northwest of Iran). European Journal of Experimental Biology. 3(1): 576-582.

11. Ghorbani, A., & A. KakehMami, 2013. Spatial database construction for natural resources and watershed management at the provincial level in Iran: A case study in Ardabil province. European Journal of Experimental Biology. 3(1): 337-347.

12. Ghorbani, A., 2015. Land use mapping and ecological capability evaluation of dry farming lands based on slope for converting to pasture in Zilbar-chay watershed using remote sensing and GIS. Journal of Geographic Space. 48: 129-149.

13. Ghorbani, A., F. Aslami, S. Ahmadabadi, & S. Gaffari, 2015. Land use mapping of Kaftareh Watershed of Ardabil using visual and digital processing of ETM+ image. Iranian Journal of Natural Ecosystems. 6(2): 133-149.

14. Google, Inc. 2007. Press Release: Introducing Google Earth Outreach, Mountain View, California, USA, 26 June, (www.google.copm/press/pressrel/outreach_20070625.html).

15. Green, E.P., P.J. Mumby, A.J. Edwards, & C.D. Clark, 2000. Remote sensing handbook for tropical coastal management. Coastal Management Sourcebooks 3, UNESCO, Paris, France: 328pp.

16. Guralnick, R.P., A.W. Hill, & M. Lane, 2007. Towards a collaborative, global infrastructure for biodiversity assessment. Journal of Ecology letters. 10(8): 663-672.

17. Hu, Q., W. Wu, T. Xia, Q. Yu, P. Yang, Z. Li, & Q. Song, 2013. Exploring the Use of Google Earth Imagery and Object-Based Methods in Land Use/Cover Mapping. Journal of Remote Sensing. 5(11): 6026-6042.

1. 18.  Jensen, J.R. 1996. Introductory Digital Image Processing, A Remote Sensing Perspective. Second edn. Prentice Hall, Saddle River, NJ, 544p.

19. Karathanassi, V., V. Andronis, & D. Rokos, 2000. Evaluation of topographic normalization methods for a Mediterranean forest area. Journal of International archives of photogrammetry and remote sensing. 33(7): 654-661.

20. Leachtenauer, J.C., K. Daniel., & T. P. Vogl. 1997. Digitizing Corona imagery: Quality vs. cost. In Corona: Between the Sun & the Earth, The first NRO reconnaissance eye in space, R.A. McDonald (ed.), American Society Photogrammetry and Remote Sensing: Washington, D.C., USA: 189-203.

21. Lefsky, M. A. & W.B. Cohen, 2003. Selection of remotely sensed data. M. A. Wulder and S. E. Franklin (eds.), remote sensing of forest environments: concepts and case studies, Kluwer Academic Publishers, Boston, USA: 13-46.

22. Lu, D., G. Li, E. Moran, C.C Freitas, L. Dutra, & S.J.S. Sant'Anna, 2012. A comparison of maximum likelihood classifier and object-based method based on multiple sensor datasets for land use/land cover classification in the Brazilian Amazon. Proceedings of the fourth GEOBIA, - Rio de Janeiro – Brazil: 7-9.

23. Macleod, R.S., & R.G. Congalton, 1998. A Quantitative Comparison of Change Detection Algorithms for Monitoring Eelegrass from Remotely Sensed Data. Journal of Photogrammetric and Engineering Remote Sensing. 64(3): 207-216.

24. Petropoulos, G. P., Ch. Kalaitzidis, & K.P. Vadrevu, 2012. Support vector machines and object-based classification for obtaining land-use/cover cartography from Hyperion hyperspectral imagery. Journal of Computers & Geosciences. 41: 99–107.

25. Rasouli, E., & H. Mohammadzadeh, 2010. Remot sensing basic of Knowledg. Elmiran publition.190pp.

26. Roostaei, S., S.A. Alavi, M.R. Nikjoo, & K.V. Kamran, 2012. Evaluation of object-oriented and pixel based classification methods for extracting changes in urban area. Journal of Geomatics and Geosciences. 2(3): 738-749.

27. Yan, G, 2003, Pixel based and object oriented image analysis for coal fire research. ITC.Fire research. 93pp.