Improving the accuracy of wetland classification based on optical and SAR imagery fusion

http://jmes.humg.edu.vn/en/archives?article=1164
  • Affiliations:

    Khoa Trắc địa - Bản đồ và Quản lý đất đai, Trường Đại học Mỏ - Địa chất, Việt Nam

  • *Corresponding:
    This email address is being protected from spambots. You need JavaScript enabled to view it.
  • Received: 15th-Mar-2017
  • Revised: 16th-June-2017
  • Accepted: 31st-Aug-2017
  • Online: 31st-Aug-2017
Views: 1390
Downloads: 598
Rating: 5.0, Total rating: 59
Yours rating

Abstract:

The pixel based spectral image analysis (PBIA) method is widely employed when mapping landscape features using information from remote-sensing imagery. However, the limitation of this method applied for mapping wetlands is low accuracy as low spectral contrast among plant species and the generally small size of vegetation zones. One of techniques can be used to improve the accuracy is multi-sensor fusion. This study is to investigate several fusion methods such as PCA, IHS, and wavelet to merge Landsat 5 TM and ERS-2, Landsat 5 TM and ALOS PALSAR, and Landsat 8 OLI and Sentinel 1. The best results are used to classify, and the hybrid fusion method of wavelet and PCA is the best one. The overall accuracy of classifying images fused increased from 2.88 to 13.09% and the Kappa coefficient increased from 0.11 to 0.14.

How to Cite
Cao, C.Xuan 2017. Improving the accuracy of wetland classification based on optical and SAR imagery fusion (in Vietnamese). Journal of Mining and Earth Sciences. 58, 4 (Aug, 2017).