Wang, Qunming and Zhang, Yihang and Onojeghuo, Alex Okiemute and Zhu, Xiaolin and Atkinson, Peter Michael (2017) Enhancing Spatio-Temporal Fusion of MODIS and Landsat Data by Incorporating 250 m MODIS Data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10 (9). pp. 4116-4123. ISSN 1939-1404
For_submission_final.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial.
Download (771kB)
Abstract
Spatio-temporal fusion of MODIS and Landsat data aims to produce new data that have simultaneously the Landsat spatial resolution and MODIS temporal resolution. It is an ill-posed problem involving large uncertainty, especially for reproduction of abrupt changes and heterogeneous landscapes. In this paper, we proposed to incorporate the freely available 250 m MODIS images into spatio-temporal fusion to increase prediction accuracy. The 250 m MODIS bands 1 and 2 are fused with 500 m MODIS bands 3-7 using the advanced area-to-point regression kriging approach. Based on a standard spatio-temporal fusion approach, the interim 250 m fused MODIS data are then downscaled to 30 m with the aid of the available 30 m Landsat data on temporally close days. The 250 m data can provide more information for the abrupt changes and heterogeneous landscapes than the original 500 m MODIS data, thus increasing the accuracy of spatio-temporal fusion predictions. The effectiveness of the proposed scheme was demonstrated using two datasets.