Journal of Marine Sciences ›› 2014, Vol. 32 ›› Issue (1): 40-46.DOI: 10.3969/j.issn.1001-909X.2014.01.005

Previous Articles     Next Articles

Comparison of coastal wetland classification accuracy using fused images of different-polarization SAR and TM images

WANG Xiao-peng1,2,3, ZHANG Jie1,3, MA Yi3, REN Guang-bo3   

  1. 1. Information Science and Technology College, Dalian Maritime University, Dalian 116026, China;
    2. Information Engineering College, Qingdao University, Qingdao 266071, China;
    3. The First Institute of Oceanography, SOA, Qingdao 266061, China
  • Received:2013-10-09 Revised:2014-02-17 Online:2014-03-15 Published:2022-11-25

Abstract: The supported vector machine was adopted on the coastal wetland land cover classification using fused images of 4-polarization SAR and TM images, the results of classification were compared and evaluated to analyze the capability of different-polarization SAR and TM fused images in coastal wetland area. An area of Yellow River Delta was selected as the experimental area, and a Radarsat-2 full-polarization image and a Landsat-5 TM image were taken as an example. The results of the experiment were analyzed quantitatively, which show that the classification accuracy of the original TM image can be improved by using the SAR-image-fusion method, the classification accuracy of the image fused by the HV-polarization SAR and TM image is the highest, so that the HV polarization is the most suitable polarization of SAR image for fusion and classification research work in coastal wetland area.

Key words: SAR, full polarization, TM, image fusion, SVM, coastal wetland, land cover

CLC Number: