An Improved FSDAF method for producing high spatiotemporal resolution NDVI

• 1、Chongqing University of Posts and Telecommunications, College of Computer Science and Technology, City Chong Qing, 400065

Abstract：Normalized difference vegetation index(NDVI)is an important way to monitor vegetation status, but due to technical and budgetary constraints, the existing global NDVI time series can not meet the requirements of high spatial and temporal resolution at the same time, that is, spatio-temporal contradiction. The Spatiotemporal fusion is an effective method to solve the contradiction of time and space. However, the existing fusion methods are mainly faced with two problems, one is to monitor catastrophe events, and the other is to maintain sufficient spatial details. To solve the above problems, a flexible spatiotemporal fusion method (FlexibleSpatiotemporalDAtaFusion,FSDAF) is proposed, but the predicted images are lack of intra-class variability and spatial details. In order to overcome the above limitations,this paper proposes an improved flexible spatio-temporal fusion method (ImprovedFlexibleSpatiotemporalDAtaFusion,IFSDAF)for generating high spatio-temporal resolution NDVI, and applies it to the image fusion of Landsat_NDVI and MODIS_NDVI. Compared with FSDAF, IFSDAF has the following advantages: (1) making full use of the information of the original data in spatial prediction; (2) introducing a new residual allocation index to improve the fusion accuracy of the abrupt region; (3) adjusting the difference between Landsat and MODIS sensors. Compared with the flexible spatiotemporal fusion model(FSDAF), the results show that the IFSDAF fusion results have better visual and quantitative effects, can more accurately reflect the spatial details of the target, and have better robustness.

Keywords： spatiotemporal fusion high spatiotemporal resolution Landsat MODIS FSDAF NDVI IFSDAF