In recent years Peter Pehani is mostly focused on automatic and semi-automatic processing of large quantities of satellite imagery. Mostly he is developing algorithms to process open access Copernicus programme data (Sentinel-2 optical images and Sentinel-1 radar images). He is also involved in different phases of processing of high and very high resolution optical data (e.g. RapidEye, SPOT and Pleiades, WordView-2/3/4, respectively), ranging from production of orthorectified images (*) to the thematic products of several thematic fields (e.g. natural hazards, forestry, statistics, land use/land cover).

He was integrator of development of the STORM – automatic near real-time satellite data processing chain, which was in early stages developed by the members of the Department of Remote Sensing ZRC SAZU under the framework of the Slovenian Centre of Excellence for Space Sciences and Technologies SPACE-SI. Over the last years development of the STORM is carried out mostly independenty of the SPACE-SI. Among several phases of STORM processing chain P.P. is primarily involved in development of submodule for image matching, i.e. matching between satellite data and vector road mask. This submodule is crucial for obtaining accurate ground control points (GCPs), used in later phases of orthorectification.

Recently he is also involved in the development of the processing workflow for NEMO-HD optical satellite data. NEMO-HD is the first Slovene satellite, operated by the SPACE-SI, launched in September 2020.

(*) Orthorectification is one of the main steps of optical satellite data processing. After orthorectification the image is properly geopositioned, and may be adequatly used with other spatial (geolocated) data. Image processed via orthorectification process is called orthorectified image or orthoimage.