Remote Sensing: Application Potentials for Sustainability and Secure Society
Pan-European Mapping of Underutilized Land for Bioenergy Production
Dr. Manuela Hirschmugl and Carina Sobe (BSc, MSc) from JOANNEUM RESEARCH – DIGITAL have published the project results of the Horizon 2020 project BIOPLAT-EU in a paper called "Pan-European Mapping of Underutilized Land for Bioenergy Production" together with colleagues from Germany and Italy.
BIOPLAT-EU aims to identify underutilized areas in Europe, which are potentially suitable for cultivating energy crops for bioenergy production using remote sensing time series analysis.
In this study, underutilized land are defined as lands that show no sign of any human use for a period of five years. The cloud-based processing platform Google Earth Engine proved to be a proper tool for efficient processing of large amounts of necessary satellite data for continental image classification. The results of this study do not only provide important information on the European-wide availability of underutilized land, that can potentially be used for the production of bioenergy resources, but also give an insight into their geographical distribution.
Highest potentials for bioenergy feedstock production were found in the Mediterranean region and in the eastern part of Europe. The classification indicates that 5.3 million ha of underutilized land in Europe is potentially available for agricultural bioenergy production.
In the next step, the data will be integrated in a webGIS platform, which allows simulating different energy crop cultivation and bioenergy pathway scenarios and enables performing ecological, economic and social sustainability assessment.
This research has received funding from the European Union’s Horizon 2020 research and innovation programme under grant No. 818083 (BIOPLAT-EU).
Critical aspects of people counting and density estimation
DI Dr. Roland Perko, Dr. Manfred Klopschitz and DI Alexander Almer from JOANNEUM RESEARCH - DIGITAL have published the results of the KIRAS projects WatchDog and KI-Secure in the article "Critical aspects of people counting and density estimation" in the MDPI Journal of Imaging together with a partner from the Technical University of Munich.
Many scientific studies deal with person counting and density estimation from single images. This work identifies limitations of the current state-of-the-art approaches and presents findings on how these limitations can be resolved. The results have shown, that the method modifications made in the study allow a significant improvement in the accuracy of people counting and density estimation compared to the state of the art. In this way, a better understanding of the underlying convolutional neural networks has been achieved. Furthermore, the research findings will be able to advance the field of person density estimation in general.