TY - JOUR A1 - Wurm, Michael A1 - Stark, Thomas A1 - Zhu, Xiao Xiang A1 - Weigand, Matthias A1 - Taubenböck, Hannes T1 - Semantic segmentation of slums in satellite images using transfer learning on fully convolutional neural networks JF - ISPRS Journal of Photogrammetry and Remote Sensing N2 - Unprecedented urbanization in particular in countries of the global south result in informal urban development processes, especially in mega cities. With an estimated 1 billion slum dwellers globally, the United Nations have made the fight against poverty the number one sustainable development goal. To provide better infrastructure and thus a better life to slum dwellers, detailed information on the spatial location and size of slums is of crucial importance. In the past, remote sensing has proven to be an extremely valuable and effective tool for mapping slums. The nature of used mapping approaches by machine learning, however, made it necessary to invest a lot of effort in training the models. Recent advances in deep learning allow for transferring trained fully convolutional networks (FCN) from one data set to another. Thus, in our study we aim at analyzing transfer learning capabilities of FCNs to slum mapping in various satellite images. A model trained on very high resolution optical satellite imagery from QuickBird is transferred to Sentinel-2 and TerraSAR-X data. While free-of-charge Sentinel-2 data is widely available, its comparably lower resolution makes slum mapping a challenging task. TerraSAR-X data on the other hand, has a higher resolution and is considered a powerful data source for intra-urban structure analysis. Due to the different image characteristics of SAR compared to optical data, however, transferring the model could not improve the performance of semantic segmentation but we observe very high accuracies for mapped slums in the optical data: QuickBird image obtains 86–88% (positive prediction value and sensitivity) and a significant increase for Sentinel-2 applying transfer learning can be observed (from 38 to 55% and from 79 to 85% for PPV and sensitivity, respectively). Using transfer learning proofs extremely valuable in retrieving information on small-scaled urban structures such as slum patches even in satellite images of decametric resolution. KW - slums KW - FCN KW - convolutional neural networks KW - deep learning KW - transfer learning Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-233799 VL - 150 ER - TY - JOUR A1 - Weigand, Matthias A1 - Wurm, Michael A1 - Dech, Stefan A1 - Taubenböck, Hannes T1 - Remote sensing in environmental justice research—a review JF - ISPRS International Journal of Geo-Information N2 - Human health is known to be affected by the physical environment. Various environmental influences have been identified to benefit or challenge people's physical condition. Their heterogeneous distribution in space results in unequal burdens depending on the place of living. In addition, since societal groups tend to also show patterns of segregation, this leads to unequal exposures depending on social status. In this context, environmental justice research examines how certain social groups are more affected by such exposures. Yet, analyses of this per se spatial phenomenon are oftentimes criticized for using “essentially aspatial” data or methods which neglect local spatial patterns by aggregating environmental conditions over large areas. Recent technological and methodological developments in satellite remote sensing have proven to provide highly detailed information on environmental conditions. This narrative review therefore discusses known influences of the urban environment on human health and presents spatial data and applications for analyzing these influences. Furthermore, it is discussed how geographic data are used in general and in the interdisciplinary research field of environmental justice in particular. These considerations include the modifiable areal unit problem and ecological fallacy. In this review we argue that modern earth observation data can represent an important data source for research on environmental justice and health. Especially due to their high level of spatial detail and the provided large-area coverage, they allow for spatially continuous description of environmental characteristics. As a future perspective, ongoing earth observation missions, as well as processing architectures, ensure data availability and applicability of ’big earth data’ for future environmental justice analyses. KW - satellite remote sensing KW - review KW - environmental justice KW - big earth data KW - urban environments Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-196950 SN - 2220-9964 VL - 8 IS - 1 ER -