Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


kent1289558937.pdf (13.04 MB)

ETD Abstract Container

Abstract Header

Object-Oriented classification of drumlins from Digital Elevation Models

Saha, Kakoli
http://rave.ohiolink.edu/etdc/view?acc_num=kent1289558937

Abstract Details

2010, PHD, Kent State University, College of Arts and Sciences / Department of Geography.
Drumlins are common elements of glaciated landscapes which are easily identified by their distinct morphometric characteristics including shape, length/width ratio, elongation ratio, and uniform direction. To date, most researchers have mapped drumlins by tracing contours on maps, or through on-screen digitization directly on top of hillshaded digital elevation models (DEMs). This paper seeks to utilize the unique morphometric characteristics of drumlins and investigates automated extraction of the landforms as objects from DEMs by Definiens Developer software (V.7), using the 30 m United States Geological Survey National Elevation Dataset DEM as input. The Chautauqua drumlin field in Pennsylvania and upstate New York, USA was chosen as a study area. As the study area is huge (approximately covers 2500 sq.km. of area), small test areas were selected for initial testing of the method. Individual polygons representing the drumlins were extracted from the elevation data set by automated recognition, using Definiens’ Multiresolution Segmentation tool, followed by rule-based classification. Subsequently parameters such as length, width and length-width ratio, perimeter and area were measured automatically. To test the accuracy of the method, a second base map was produced by manual on-screen digitization of drumlins from topographic maps and the same morphometric parameters were extracted from the mapped landforms using Definiens Developer. Statistical comparison showed a high agreement between the two methods confirming that object-oriented classification for extraction of drumlins can be used for mapping these landforms. The proposed method represents an attempt to solve the problem by providing a generalized rule-set for mass extraction of drumlins. To check that the automated extraction process was next applied to a larger area. Results showed that the proposed method is as successful for the bigger area as it was for the smaller test areas.
Dr. Mandy Munro-Stasiuk (Committee Chair)
Dr. Scott Sheridan (Committee Member)
Dr. Jay Lee (Committee Member)
Dr. Abdul Shakoor (Committee Member)
Dr. Cheng- Chang Lu (Committee Member)
148 p.
Geography
Object-orineted classification; DEMs; Drumlins

Recommended Citations

Citations

  • Saha, K. (2010). Object-Oriented classification of drumlins from Digital Elevation Models [Doctoral dissertation, Kent State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=kent1289558937

    APA Style (7th edition)

  • Saha, Kakoli. Object-Oriented classification of drumlins from Digital Elevation Models. 2010. Kent State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=kent1289558937.

    MLA Style (8th edition)

  • Saha, Kakoli. "Object-Oriented classification of drumlins from Digital Elevation Models." Doctoral dissertation, Kent State University, 2010. http://rave.ohiolink.edu/etdc/view?acc_num=kent1289558937

    Chicago Manual of Style (17th edition)

kent1289558937
1,433
© 2010, all rights reserved.
This open access ETD is published by Kent State University and OhioLINK.
Release 3.2.12