Suchergebnisse

Land cover Classified maps obtained from deep learning methods such as Convolutional neural networks (CNNs) and fully convolutional networks (FCNs) usually have high classification accuracy but with the detailed structures of objects lost or smoothed. In this work, we develop a methodology based on fully convolutional networks (FCN) that is trained in an end-to-end fashion using aerial RGB images only as input. Skip connections are introduced into the FCN architecture to recover high spatial details from the lower convolutional layers. The experiments are conducted on the city of Goma in the Democratic Republic of Congo. We compare the results to a state-of-the art approach based on a semi-automatic Geographic object image-based analysis (GEOBIA) processing chain. State-of-the art classification accuracies are obtained by both methods whereby FCN and the best baseline method have an overall accuracy of 91.3% and 89.5% respectively. The maps have good visual quality and the use of an FCN skip architecture minimizes the rounded edges that is characteristic of FCN maps. Additional experiments are done to refine FCN classified maps using segments obtained from GEOBIA generated at different scale and minimum segment size. High OA of up to 91.5% is achieved accompanied with an improved edge delineation in the FCN maps, and future work will involve explicitly incorporating boundary information from the GEOBIA segmentation into the FCN pipeline in an end-to-end fashion. Finally, we observe that FCN has a lower computational cost than the standard patch-based CNN approach especially at inference. ; SCOPUS: ar.j ; info:eu-repo/semantics/published

Remote sensing images provide the capability to obtain information of various landcover classes. This information is useful to the process of urban planning, socio-economicmodelling and population studies. There is a need for accurate and efficient methods to obtainthis information, particularly in regions where there is a scarcity of reference data. In thiswork, we develop a methodology based on fully convolutional networks (FCN) that is trainedin an end-to-end fashion using aerial RGB images only as input. The experiments are conductedon the city of Goma in the Democratic Republic of Congo. We compare the results to a state-of-the art approach based on a semi-automatic Geographic object image-based analysis(GEOBIA) processing chain. State-of-the-art classification accuracies are obtained by bothmethods whereby FCN and the best baseline method have an overall accuracy of 91.24% and89.34% respectively. The maps have good visual quality and the use of a FCN skip architectureminimizes the rounded edges that is characteristic of FCN maps. Finally, additionalexperiments are done to refine FCN classified maps using segments obtained from GEOBIA.This resulted in improved edge delineation in the FCN maps, and future work will involveexplicitly incorporating boundary information from the GEOBIA segmentation into the FCNpipeline in an end-to-end fashion. ; info:eu-repo/semantics/published

Routine and accurate data on deprivation are needed for urban planning and decision support at various scales (i.e. from community to international). However, analyzing information requirements of diverse users on urban deprivation, we found that data are often not available or inaccessible. To bridge this data gap, Earth Observation (EO) data can support access to frequently updated spatial information. However, a user-centered approach is urgently required for the production of EO-based mapping products. Combining an online survey and several forms of user interactions, we defined five system specifications (derived from user requirements) for designing an open-access spatial information system for deprived urban areas. First, gridded maps represent the optimal spatial granularity to deal with high uncertainties of boundaries of deprived areas and to protect privacy. Second, a high temporal granularity of 1–2 years is important to respond to the high spatial dynamics of urban areas. Third, detailed local-scale information should be part of a city-to-global information system. Fourth, both aspects, community assets and risks, need to be part of an information system, and such data need to be combined with local community-based information. Fifth, in particular, civil society and government users should have fair access to data that bridges the digital barriers. A data ecosystem on urban deprivation meeting these requirements will be able to support community-level action for improving living conditions in deprived areas, local science-based policymaking, and tracking progress towards global targets such as the SDGs. ; info:eu-repo/semantics/published