Classification of forest areas by k nearest neighbor method: case study, Antalya

Download

index.pdf

Date

2008

Author

Özsakabaşı, Feray

Metadata

Show full item record

Item Usage Stats

130
views

42
downloads

Among the various remote sensing methods that can be used to map forest areas, the K Nearest Neighbor (KNN) supervised classification method is becoming increasingly popular for creating forest inventories in some countries. In this study, the utility of the KNN algorithm is evaluated for forest/non-forest/water stratification. Antalya is selected as the study area. The data used are composed of Landsat TM and Landsat ETM satellite images, acquired in 1987 and 2002, respectively, SRTM 90 meters digital elevation model (DEM) and land use data from the year 2003. The accuracies of different modifications of the KNN algorithm are evaluated using Leave One Out, which is a special case of K-fold cross-validation, and traditional accuracy assessment using error matrices. The best parameters are found to be Euclidean distance metric, inverse distance weighting, and k equal to 14, while using bands 4, 3 and 2. With these parameters, the cross-validation error is 0.009174, and the overall accuracy is around 86%. The results are compared with those from the Maximum Likelihood algorithm. KNN results are found to be accurate enough for practical applicability of this method for mapping forest areas.

Subject Keywords

Geodesy.

URI

http://etd.lib.metu.edu.tr/upload/12609548/index.pdf
https://hdl.handle.net/11511/18232

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Classification of remotely sensed data by using 2D local discriminant bases Tekinay, Çağrı; Çetin, Yasemin; Department of Information Systems (2009) In this thesis, 2D Local Discriminant Bases (LDB) algorithm is used to 2D search structure to classify remotely sensed data. 2D Linear Discriminant Analysis (LDA) method is converted into an M-ary classifier by combining majority voting principle and linear distance parameters. The feature extraction algorithm extracts the relevant features by removing the irrelevant ones and/or combining the ones which do not represent supplemental information on their own. The algorithm is implemented on a remotely sensed...
Different orbit determination algorithms for BİLSAT-1 Ural, Serkan; Karslıoğlu, Mahmut Onur; Department of Geodetic and Geographical Information Technologies (2006) This study aims to investigate different orbit determination algorithms for the first Turkish remote sensing satellite, BILSAT-1. The micro-satellite carries an onboard GPS receiver. Pseudorange measurements simulated from the position and velocity data supplied by TÜBITAK-BILTEN are used for the implementation of different orbit determination algorithms concluding to an estimate of the satellite̕s state. Satellite̕s position, velocity components and the GPS receiver̕s clock bias are selected as the state p...
Nonblocking hierarchical control of decentralized des Schmidt, Klaus Verner; Moor, Thomas (null; 2005-12-01) This work considers a hierarchical control architecture for a class of discrete event systems which can also be applied to decentralized control systems. It is shown that nonblocking supervisory control on the high level of the hierarchy results in nonblocking and hierarchically consistent control on the low level. Copyright © 2005 IFAC.
Tracking of ground targets with interacting multiple model estimator Acar, Duygu; Baykal, Buyurman; Department of Electrical and Electronics Engineering (2011) Interacting Multiple Model (IMM) estimator is used extensively to estimate trajectories of maneuvering targets in cluttered environment. In the standard tracking methods, it is assumed that movement of target is applicable to a certain model and the target could be monitored via the usage of status predictions of that model. However, targets can make different maneuvering movements. At that time, expression of target dynamic model with only one model can be insufficient. In IMM approach, target dynamic mode...
Hierarchical discrete event systems with inputs and outputs Perk, Sebastian; Moor, Thomas; Schmidt, Klaus Verner (null; 2006-12-01) We propose a framework for the hierarchical design of discrete event systems that addresses both safety and liveness properties. Technically, we build on a notion of inputs and outputs that is closely related to J.C. Willems' behavioural systems theory. We develop a structural admissibility condition that allows for abstraction-based controller synthesis similar to previous work on hybrid control systems. A key feature of our framework is an alternation of subsystem composition and controller synthesis that...

Citation Formats

F. Özsakabaşı, “Classification of forest areas by k nearest neighbor method: case study, Antalya,” M.S. - Master of Science, Middle East Technical University, 2008.