Please use this identifier to cite or link to this item: http://localhost:8081/jspui/handle/123456789/16615
Title: MODEL BASED DECOMPOSITION OF POLSAR IMAGES
Authors: Kumar, Abhishek
Keywords: Terrain Classification;Radar line;Report;Orientation Angle
Issue Date: May-2017
Publisher: I I T ROORKEE
Abstract: For terrain classification and target decomposition, deorientation playes a main role in PolSAR target decomposition. The present method of deorientation roughly rotates the target about the radar line of sight by an average orientation angle(OA) in model based decomposition of PolSar. But the case when the target is complex type(high entropy mixed scatterers with different orientation angles), The existing deorientation technique becomes insufficient to handle this diverse orientation angles situation. To address the above problem, a new deorientation method is proposed in this report. The given deorientation method rotates the target coherency and matrix depending upon the dominent sacttering. This method also deals with to minimize the negative power problem .This method provides a better deorientation of the man-made type of structure hence it is good for extraction of the oriented urban reasons. A comparison with the RADARSAT-2 PolSAR data demonstrate the better deorientation performance of the proposed method.
URI: http://localhost:8081/jspui/handle/123456789/16615
metadata.dc.type: Other
Appears in Collections:MASTERS' THESES (E & C)

Files in This Item:
File Description SizeFormat 
G27531.pdf5.26 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.