Remote Sensing Based Building Extraction II

<p>Building extraction from remote sensing data plays an important role in geospatial applications such as urban planning disaster management navigation and updating geographic databases. The rapid development of image processing techniques and the accessibility of very-high-resolution multispectral hyperspectral LiDAR and SAR remote sensing images have further boosted research on building-extraction-related topics. In particular to meet the recent demand for advanced artificial intelligence models many research institutes and associations have provided open source datasets and annotated training data presenting new opportunities to develop advanced approaches for building extraction and monitoring. Hence there are higher expectations of the efficiency accuracy and robustness of building extraction approaches. Additionally they should meet the demand for processing large city- national- and global-scale datasets. Moreover learning and dealing with imperfect training data remains a challenge as does unexpected objects in urban scenes such as trees clouds and shadows. In addition to building masks more research has arisen on the automatic generation of LoD2/3 building models from remote sensing data. This follow-up Special Issue of Remote Sensing-based Building Extraction has collected more research on cutting-edge approaches to essential urban processes such as 3D reconstruction automatic building segmentation and 3D roof modelling.</p>