Abstract
Abstract. Accurate and complete geographic data of human settlements is crucial for effective emergency response, humanitarian aid and sustainable development. Open- StreetMap (OSM) can serve as a valuable source of this data. As there are still many areas missing in OSM, deep neural networks have been trained to detect such areas from satellite imagery. However, in regions where little or no training data is available, training networks is problematic. In this study, we proposed a method of transferring a building detection model, which was previously trained in an area wellmapped in OSM, to remote data-scarce areas. The transferring was achieved via fine-tuning the model on limited training samples from the original training area and the target area. We validated the method by transferring deep neural networks trained in Tanzania to a site in Cameroon with straight distance of over 2600 km, and tested multiple variants of the proposed method. Finally, we applied the fine-tuned model to detect 1192 buildings missing OSM in a selected area in Cameroon. The results showed that the proposed method led to a significant improvement in f1-score with as little as 30 training examples from the target area. This is a crucial quality of the proposed method as it allows to fine-tune models to regions where OSM data is scarce.
Highlights
Spatial data are fundamental for long-term planning and sustainable development as well as immediate emergency and disaster response
We proposed a method of transferring a building detection model, which was previously trained in an area wellmapped in OSM, to remote data-scarce areas
We proposed a novel method of detecting buildings missing in OSM by transferring deep neural networks to geographically remote regions with a limited amount of additional training data
Summary
Spatial data are fundamental for long-term planning and sustainable development as well as immediate emergency and disaster response. Herfort et al (2019) proposed a method of training deep neural networks to detect buildings from satellite imagery and assist another crowdsourced mapping platform, namely MapSwipe, to better and faster identify human settlements which are missing in OSM. A similar approach to (Li et al, 2019) was applied, using pretrained deep networks to classify image tiles with high probability of OSM missing built-up areas. Both studies trained their networks in areas where OSM coverage of buildings was relatively high and applied their networks in relative proximity to where the networks were trained. These differences include diverse landscapes, sources of satellite imagery or appearances of buildings
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
