Optimal Collection of High Resolution Aerial Imagery with Unmanned Aerial Systems

Remote sensing applications are an emerging topic for Unmanned Aerial Systems (UASs). Unlike many remote sensing image collection methods, UASs have several advantages when it comes to on demand data acquisition. Relatively low operating costs, high portability and low flight altitudes make UASs excellent tools for researchers to collect high resolution imagery where satellites or manned aircraft are inefficient. In particular areas, such as in rangelands, the use of UASs to aid in management practices could have significant benefit. However, in these areas, current methodologies of remote sensing utilizing spectral reflectance data for vegetation analysis have performed poorly due to the high spatial and low spectral heterogeneity of the area. One of the root causes of the poor performance can be traced to the negative effect of shadows that are interspersed in the spectral reflectance data. The unique advantage of low infrastructure and minimal downtime for UASs enables researchers to exert greater control over the precise time of data collection. In this paper, it is demonstrated that the time of imagery collection can be optimized with regards to the minimization of shadows found in the imagery. The process described in this paper utilizes a high resolution digital elevation map (DEM) that can be generated through photogrammetry techniques to create an estimate of shadows given a time of day at a known location. Furthermore, the results of estimated shadow map can be utilized for improving classification techniques without additional equipment.