Normalization method for multi-sensor high spatial and temporal resolution satellite imagery with radiometric inconsistencies

Abstract

Distributed systems of small satellites are generating a new type of remote sensing data: multi-spectral data with high spatial and temporal resolution and near-daily global coverage. This data is proving to be valuable for monitoring land cover change. However, in order to achieve widespread application for this purpose, all satellites in the distributed system, or constellation, must acquire imagery with accurate georeferencing and consistent radiometric properties. In this research, we developed a method to automatically co-register and radiometrically normalize a temporally dense series of smallsat images using geocorrected reference imagery. To demonstrate the approach, we normalized a smallsat image time series at both the pixel and the polygon level, smoothing spectral indices through time to detect both abrupt and gradual changes. Using PlanetScope imagery, we tested these methods in a forested region of British Columbia, Canada heavily impacted by forest fires in 2017. By examining the normalized difference vegetation index (NDVI) in the acquired imagery before and after the fires, we found that this method allowed simple identification of burned and unburned areas, which was not readily possible without applying the normalization method. Our result suggests that the developed approach can help fully exploit remote sensing datasets that have high spatial resolution and are acquired with high frequency but potentially contain radiometric inconsistencies in order to quickly identify land cover changes.