Title :
Autumn crop identification using high-spatial-temporal resolution time series data generated by MODIS and Landsat remote sensing images
Author :
Dengfeng Xie ; Peijun Sun ; Jinshui Zhang ; Xiufang Zhu ; Wenna Wang ; Zhoumiqi Yuan
Author_Institution :
State Key Lab. of Earth Surface Processes & Resource Ecology, Beijing Normal Univ., Beijing, China
Abstract :
Toward the problems existing in the fragmented land parcel of autumn crops in china and the low resolution of MODIS and difficulty in obtaining high-quality Landsat data, spatial and temporal fusion technology with MODIS and Landsat data is used to generate high-spatial-temporal resolution time series data. Then, combine the data generated and use SVM method to identify autumn crops and then determine which data to be the best choice for autumn crop identification. The results show the overall accuracy of paddy and corn with red data is the highest up to 87.47% and the user´s accuracy of paddy and corn of NDVI data reached 84.91% and 75.36%, respectively. Overall it is feasible to use the high-spatial-temporal resolution time series data generated to identify autumn crops.
Keywords :
geophysical image processing; geophysical techniques; image fusion; remote sensing; vegetation; China; LANDSAT remote sensing image; Landsat data; MODIS data; MODIS low resolution; MODIS remote sensing image; NDVI corn data; NDVI paddy data; autumn crop identification; autumn crops; fragmented land parcel; high-quality Landsat data; high-spatial-temporal resolution time series data; spatial fusion technology; temporal fusion technology; Accuracy; Agriculture; Earth; Image resolution; MODIS; Remote sensing; Satellites; Landsat; MODIS; autumn crop identification; spatial and temporal fusion technology; time series data;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2014 IEEE International
Conference_Location :
Quebec City, QC
DOI :
10.1109/IGARSS.2014.6946884
