Support Vector Machine (SVM) is characteristic of processing complex data and high dimensional data. In this paper, a new approach of image decision-level fusion based on SVM and the corresponding fusion rule based on consensus theoretic were presented. Then to select a test area in Shaoxing City, Zhejiang Province, China, a fusion experiment was conducted using Landsat TM multispectral data (30 m) and IRS-C Pan data (5.8 m). Finally an evaluation on the fusion image was given. The results show that the overall classification accuracy of the fusion image reached 81.05%. The new fusion method could satisfy the requirement of land cover classification automatically.
A new method of remote sensing image decision-level fusion based on Support Vector Machine
2003-01-01
338734 byte
Conference paper
Electronic Resource
English
Collecting Decision Support System Data through Remote Sensing of Unpaved Roads
Transportation Research Record | 2014
|