Title :
Regional land cover classification from MODIS time-series and geographical data using support vetor machine
Author :
Cai, Hongyan ; Zhang, Shuwen
Author_Institution :
Northeast Inst. of Geogr. & Agroecology, Changchun, China
Abstract :
The study investigated the performance of support vector machine (SVM) classifier for regional land cover mapping. First, 8 input features derived from MODIS time series and DEM data were selected by Jeffreys-Matusita distance. Then, all the features were analyzed to generate land cover map of Sanjiang Plain in China, using SVM algorithm. Finally, we evaluated the impact of sample size and its distribution on classification accuracy. The train and test ratio of 8:2 was proved to be a better choice for improving land cover classification. The distribution of samples influenced classification results, with a standard deviation of 0.81 to overall accuracy and 0.01 to Kappa coefficient. The overall accuracy of resultant classification map was 96.45 with Kappa coefficient of 95.8%. The good performance indicated great potentials of SVM algorithm for regional land cover mapping.
Keywords :
classification; geographic information systems; regional planning; support vector machines; terrain mapping; time series; China; DEM data; Jeffreys-Matusita distance; Kappa coefficient; MODIS time-series; Sanjiang Plain; classification; geographical data; regional land cover mapping; standard deviation; support vector machine; Accuracy; Classification algorithms; MODIS; Remote sensing; Support vector machines; Time series analysis; Training; Feature extraction and selection; Land cover classification; MODIS; Sample size; Support vector machine;
Conference_Titel :
Information Computing and Telecommunications (YC-ICT), 2010 IEEE Youth Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-8883-4
DOI :
10.1109/YCICT.2010.5713055
