Spatio-Temporal image fusion using landsat-8 and himawari-8 satellite imagery

Satellite remote sensing images provide periodical and multispectral information about the ground surface, which empowers various monitoring applications. Among the applications, some of them requires high spatial and temporal resolution, including disaster management, crop phenology monitoring, land cover change detection, etc. In order to provide high spatial and temporal information, image fusion techniques were proposed to integrate high-spatial-resolution satellite imagery with high-Temporal-resolution imagery, such as the spatial and temporal adaptive reflectance fusion model (STARFM). Previous studies applied the STARFM method on Landsat and Moderate-resolution Imaging Spectroradiometer (MODIS) satellite images to provide high spatial and high temporal images. With the advance of satellite technology, a new satellite, Himawari-8, has a 10-minute temporal resolution, which is much higher than that of MODIS, and its spatial resolution is similar to MODIS. This research tries to fuse the Landsat-8 and Himawari-8 images to provide images with high spatial resolution and very high temporal resolution. On the preprocessing step, we use affine transformation for geometric correction. Then we use the STARFM method to fuse Landsat-8 images with Himawari-8 images to produce 30-meter-spatial-resolution and 10-minute-Temporal-resolution images. Finally, in order to verify the result of this research, we apply the fused images to estimate aerosol optical depth (AOD) and compare with in-situ observations. Overall, this research uses Himawari- 8 and Landsat-8 image to produce images with high spatial and temporal resolutions. The fused images could provide timely information to assist various monitoring applications.