Publications

Tu, CM; Li, F (2019). Type Classification of Agricultural Cultivated Land Using Remote Sensing. REVISTA DE LA FACULTAD DE AGRONOMIA DE LA UNIVERSIDAD DEL ZULIA, 36(4), 1193-1202.

Abstract
As the level of agricultural cultivation continues to rise, more and more operational techniques are applied to agricultural arable land. The classification of agricultural land in the macro is still a big problem. In order to solve the problem of inaccurate classification of agricultural land, this paper uses remote sensing technology to add more spatial details provided by Landsat images. The fused NDVI time series has better distinguishing ability for various wetland types, and the improved spatial resolution also makes the fused time series is better able to capture changes in the environmental gradient. Compared with MODIS-NTS, the time series data with 30 m spatial resolution is significantly improved (p value <0.05) to improve the recognition of floating vegetation and submerged vegetation. Overall, as the spatial resolution decreases, the area-weighted average shape factor (AWMSI) shows a downward trend. The larger the value, the more complex and irregular the plaque shape is, the smaller the value, the simpler the shape. The change ratio of each category of the change of the classification result of the multi-training sample before and after the fusion is smaller than that of the multi-classifier before and after the fusion, and the changes of the classification before and after the multi-sample set classification result are also different. In terms of the number of pixels, the change of different categories has a certain relationship with the area base of the category itself.

DOI:

ISSN:
0378-7818