2:15 PM - 2:30 PM
[U05-08] Identification parameters affecting automated georeferencing accuracy of Earth observation microsatellite's multispectral images
★Invited Papers
Keywords:Georeferencing, MMSATS-1 microsatellite, Image matching, Spaceborne multispectral imager(SMI)
As microsatellites are low-cost and high performance in remote sensing, they are become suitable to utilize in Earth observation missions such as the monitoring of glacier-melting, climate change process, water availability and coastal modifications, vegetation changes monitoring, deforestation floods, disaster monitoring, border surveillance and so on. The remote sensing data which is provided from microsatellites is becoming more useful for earth observation day by day. Therefore, the preprocessing corrections for microsatellite’s images are important to apply images accurately for related fields. One of preprocessing corrections, georeferencing means a process to relate the internal coordinate system of a map or aerial photo image to a geographic coordinate system. Without location information, the raster images cannot be analyzed and compared pixel by pixel with other spatial data. Besides, wrong georeferencing leads to inaccurate map projections and measurements that require post-processing correction. Georeferencing accuracy of microsatellite is not perfect since the error in attitude measurements and satellite position remains because of their limited sensor’s performance. Therefore, the georeferencing algorithm based on image matching without using payload sensor’s data have been developed. However, image matching process for large size and scene remote sensing images can be difficult depending on various conditions such as spectrum bands, sensor, time and geometry. Therefore, the objectives of this research are to identify important geophysical parameters (brightness, viewing angle, local solar time, etc.) affecting the georeferencing accuracy and to clarify geographical and spectral dependence of the georeferencing accuracy.
In this research, at first, the automated georeferencing algorithm for MMSATS-1 microsatellite’s multispectral images was established using image matching. In this matching process, MMSATS-1 microsatellite’s spaceborne multispectral imager (SMI) images are used as query image which is necessary to be georeferenced and Sentinel-2 multispectral images are applied as base image which is already georeferenced. After developing automated georeferencing algorithm, root mean square error (RMSE) was calculated to check georeferencing accuracy. And then, the geophysical parameters affecting accuracy were identified by using different ways. As the results, this research has shown that the georeferencing accuracy depends on the parameters such as atmospheric scattering effect, brightness of SMI image, viewing angle, temporal difference between NIR band of SMI image and Sentinel-2 image. Therefore, regarding these results will be presented and discussed in this meeting.
In this research, at first, the automated georeferencing algorithm for MMSATS-1 microsatellite’s multispectral images was established using image matching. In this matching process, MMSATS-1 microsatellite’s spaceborne multispectral imager (SMI) images are used as query image which is necessary to be georeferenced and Sentinel-2 multispectral images are applied as base image which is already georeferenced. After developing automated georeferencing algorithm, root mean square error (RMSE) was calculated to check georeferencing accuracy. And then, the geophysical parameters affecting accuracy were identified by using different ways. As the results, this research has shown that the georeferencing accuracy depends on the parameters such as atmospheric scattering effect, brightness of SMI image, viewing angle, temporal difference between NIR band of SMI image and Sentinel-2 image. Therefore, regarding these results will be presented and discussed in this meeting.