Break Detection for Satellite Time Series Data

Large amounts of satellite data are now becoming available, which, in combination with appropriate change detection methods, offer the opportunity to derive accurate information on timing and location of disturbances such as deforestation events across the earth surface. Typical scenarios require the analysis of billions of image patches/pixels, which can become very expensive from a computational point of view. The bfast package provides an efficient massively-parallel implementation for one of the state-of-the-art change detection methods called Breaks For Additive Season and Trend (BFASTmonitor) proposed by Verbesselt et al.

The implementation is based on OpenCL and allows to process large-scale change detection scenarios given satellite time series data. The optimizations made are tailored to the specific requirements of modern massively-parallel devices such as graphics processing units (GPUs). The image below shows the output of the massively-parallel implementation provided by this package applied to satellite time series data covering the entire continental tropical Africa. It illustrates the detected breaks with a negative magnitude (red). The inset figures zoom in on deforestation areas. A1 and B1 emphasize the timing of the detected breaks and A2 and B2 show the deforestation on RGB satellite imagery, respectively.

alternate text

Application of BFAST-Monitor

The source code is published under the GNU General Public License (GPLv3).