Fast & Furious: Fourier transforms applied to the estimate of multi-temporal landscape fragmentation

Landscape fragmentation is known to be related to ecosystem degradation. Multi-temporal analysis based on remotely sensed data has played an important role in detecting fragmentation, typically by creating land use maps from remotely sensed images acquired at different dates. However, general classification may present a number of drawbacks such as: i) an implicit degradation of the information content of images due to classification, coupled with ii) the loss of the continuous information about fragmentation processes, and iii) problems related to downscaling once images are classified with different resolutions. Alternative approaches based on continuous information for detecting fragmentation are thus strongly encouraged. The aim of this study is to describe a Fourier transform Open Source tool to detect potential fragmentation over the landscape.