Tag: machine-learning
-
Paper on Analysis of Feature Relevance in Deep Learning for 3D Point Cloud Classification
A paper investigating the relevance of (pre-calculated) features for 3D point cloud classification using deep learning was just published in the ISPRS Annals of Photogrammetry and Remote Sensing. The study presents a non-end-to-end deep learning classifier for 3D point clouds using multiple sets of input features and compares it with an implementation of the state-of-the-art…
-
Disaster Risk Reduction, OpenStreetMap and Missing Maps at Global Platform 2019
Global Platform 2019 in Geneva Creating maps helps humanity. Drawing maps together with communities is crucial for effective risk reduction interventions, ensuring no one is left behind. The progress of the implementation of the targets set by the Sendai Framework for Disaster Risk Reduction (DRR) have been key discussion points during this years Global Platform…
-
HeiGIT supports Citizien Science @ Helmholtz ThinkCamp
We are always happy to support citizen science projects at the HeiGIT. HeiGIT/ GIScience efforts already range from tools that assess the data quality of citizen science projects (see, e.g., this blog post about “Plausible Parrots“) to approaches related to data creation, like MapSwipe Analytics (learn more here). Currently, we are supporting citizen science approaches…
-
Machine Learning for Space and Earth Observation Data (ML-SEOD) 2019 at ICCSA
Machine Learning for Space and Earth Observation Data (ML-SEOD) 2019 Call for Papers The Earth and Space environments are being monitored by an unprecedented amount of sensors: Earth observation satellites, sensor networks, telescopes working in different wavelengths, human records of Earth and Space events, etc. This generates a huge amount of raw data that must…
-
Successfull PhD Defence of Michael Schultz about tropical deforestation monitoring using Landsat time series and breakpoint detection
Last week our team member Dr. Michael Schultz successfully defended his PhD Defence about Tropical deforestation monitoring using Landsat time series and breakpoint detection at Wageningen University and Research (WUR). It was supervised by Prof. Dr M. Herold (Professor of Geo-information Science and Remote Sensing at Wageningen University & Research). We congratulate cordially! Keep up…
-
Deep Learning from Multiple Crowds: A Case Study of Humanitarian Mapping
Our paper about Deep Learning from Multiple Crowds: A Case Study of Humanitarian Mapping is available online now. Satellite images are widely applied in humanitarian mapping which labels buildings, roads and so on for humanitarian aid and economic development. However, the labeling now is mostly done by volunteers. In a recently accepted study, we utilize deep…
-
Keynote on VGI and AI for Disaster Management by Prof. Zipf at 10 Years HAUS New York
Heidelberg Alumni USA (HAUS) is the alumni initiative for all former students, lecturers, professors or staff of Heidelberg University who live or work in the US. HAUS was born in New York City, and that’s where we will celebrate the tenth anniversary the weekend of September 14-16, 2018. We hope you will be there for…
-
EU Sentinel 2 osmlanduse.org fusion @ Toulouse Space show
Another chapter in machine human fusion land use device narrative: new Sentinel 2 osmlanduse.org product results based on OpenStreetMap plus Sentinel 2 data plus Machine Learning were presented at ToulouseSpaceShow 2018 during a European Space Agency (ESA) Research and User Support (RUS) event. Stay tuned: The new product will soon be available for all EU…
-
OSMlanduse.org + Remote Sensing = New conterminous land use data set for Germany released – filling gaps in OSM through machine learning
Land use data created by humans (OSM) was fused with satellite remote sensing data, resulting in a conterminous land use data set without gaps. The first version is now available for all Germany at OSMlanduse.org. When human input (OSM data) was absent a machine generated missing land use information learning from human inputs and using…
-
LandSense is Project of the Week @ “Doing It Together Science”.
The EU H2020 project LandSense (A Citizen Observatory and Innovation Marketplace for Land Use and Land Cover Monitoring) has been featured as project of the week by “Doing It Together Science“. http://togetherscience.eu/blog/project-of-the-week-10-landsense. Thanks to our partners at IASA etc. for this! In addition to organizing mapathons and related research research activities, in the context of…
-
Job Offer: Software Engineer Geoinformation Technology / OpenStreetMap
Open Position: Software Engineer Geoinformation Technology / OpenStreetMap Heidelberg Institute for Geoinformation Technology (HeiGIT) In order to promote technology transfer and applied research in the area of Geoinformatics, the Heidelberg Institute for Geoinformation Technology (HeiGIT) is currently being established with the support of the Klaus-Tschira Foundation. http://www.heigit.org. This is to be continued in the future…
-
GIScience Colloquium talk on “Heatmapping”: Accessing Geodatabases of Building Stocks for the Development of Spatial Energy System Models by Sebastian Blömer
we cordially invite everybody interested to our next open GIScience colloquium talk “Heatmapping”: Accessing Geodatabases of Building Stocks for the Development of Spatial Energy System Models Sebastian Blömer ifeu Institut für Energie- und Umweltforschung Heidelberg Time and date: Mon, July 10, 2.15 pm Venue: INF 348, Room 015, Department of Geography, Heidelberg University To evaluate the…