Following up on last month’s piece on vegetation analysis utilizing 5-meter RapidEye imagery, for the March edition of The Geospatial Times we will demonstrate the value of this data for extracting building footprints across regional landscapes. RapidEye is a constellation of five 5-meter medium resolution satellites each offering 5 spectral bands of information. The RapidEye constellation offers a daily revisit time to every location on the planet with a huge footprint that is 40-km wide. The data is priced extremely competitively with a base price of $1.28 per square kilometer for all 5 spectral bands – academics receive a discount on this price. RapidEye adds a fifth band, the red edge, to the ‘traditional’ multispectral set of blue, green, red and near-infrared (NIR). The additional spectral data available in the red edge band allows users to extract more useful ‘information’ than you can from traditional 4-band imagery sources.
While RapidEye has lower resolution than data offered by DigitalGlobe and GeoEye, its wide area coverage and red edge band make it an ideal solution for extracting building footprints from a large region. To create the animation below, we used ENVI to extract the footprints of buildings with bright ‘white’ roofs in an automated fashion from a scene of RapidEye data collected over the San Luis Valley, Colorado. To complete this extraction, we first built a training set of white roofs utilizing the Region of Interest (ROI) tool in ENVI 4.8. This training set contained ten polygons (approximately 375 pixels) of evenly dispersed white roofs throughout the RapidEye scene. Next, we ran a Parallelepiped Supervised Classification filter set to identify pixels in the imagery that were within 3 standard deviations of the mean spectral value of the ROI.
A quick inspection of the results revealed that we had room for improvement so an extra polygon over a missed white roof was added into the ROI. The same Parallelepiped filter was run on the entire RapidEye scene again yielding the results you see below. While we do not purport these results to be perfect, they are quite accurate especially given the limited tweaking of the ROI and filtering techniques we employed in this brief analysis. By adding several additional polygons to the ROI as well as tweaking the supervised classification filter, we are confident that these results would be even more accurate.
If you would like to find out more about using RapidEye for your academic, engineering or any landscape study, let us know at [email protected] or (303) 993-3863.