Abstract
Abstract. Deep learning methods based on Fully convolution networks (FCNs) have shown an impressive progress in building outline delineation from very high resolution (VHR) remote sensing (RS) imagery. Common issues still exist in extracting precise building shapes and outlines, often resulting in irregular edges and over smoothed corners. In this paper, we use PolyMapper, a recently introduced deep-learning framework that is able to predict object outlines in a vector representation directly. We have introduced two main modifications to this baseline method. First, we introduce EffcientNet as backbone feature encoder to our network, which uses compound coefficient to scale up all dimensions of depth/width/resolution uniformly, to improve the processing speed with fewer parameters. Second, we integrate a boundary refinement block (BRB) to strengthen the boundary feature learning and to further improve the accuracy of corner prediction. The results demonstrate that the end-to-end learnable model is capable of delineating polygons of building outlines that closely approximate the structure of reference labels. Experiments on the crowdAI building instance segmentation datasets show that our model outperforms PolyMapper in all COCO metrics, for instance showing a 0.13 higher mean Average Precision (AP) value and a 0.60 higher mean Average Recall value. Also qualitative results show that our method segments building instances of various shapes more accurately.
Highlights
Automatic building extraction from remote sensing images has been a core research topic in the remote sensing area for decades
In contrast with standard vector labels and georegistered image files, each object instance annotation in Common Objects in Context (COCO) .json contains a series of fields, including the category id, segmentation mask, and enclosing bounding box coordinates of the object
We compared our model to the state-of-the-art instance segmentation method Mask R-convolutional neural networks (CNNs) (He et al, 2017) and the original PolyMapper
Summary
Automatic building extraction from remote sensing images has been a core research topic in the remote sensing area for decades It has many applications, including cadastral and topographic mapping, cartography, urban planning, and humanitarian aid. Zhao et al, (2018) applied the Mask R-CNN for building segmentation and a regularization algorithm to polygonize segmentation results Those methods are either not end-to-end trainable or only capable of handling building objects with simple shapes. There are two main contributions in this study: 1) Introduction of a state-of-the-art architecture for the backbone network, EfficientNet (Tan et al, 2019), which uses a compound coefficient to scale up CNNs in a more structured manner This reduces the number of trainable parameters while maintaining high accuracy.
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 callMore From: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: 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.
