Abstract
The lack of pixel-level labeling limits the practicality of deep learning-based building semantic segmentation. Weakly supervised semantic segmentation based on image-level labeling results in incomplete object regions and missing boundary information. This paper proposes a weakly supervised semantic segmentation method for building detection. The proposed method takes the image-level label as supervision information in a classification network that combines superpixel pooling and multi-scale feature fusion structures. The main advantage of the proposed strategy is its ability to improve the intactness and boundary accuracy of a detected building. Our method achieves impressive results on two 2D semantic labeling datasets, which outperform some competing weakly supervised methods and are close to the result of the fully supervised method.
Highlights
Building detection plays an important role in urban development planning, urban infrastructure planning, urban land use and management, land use change monitoring, digital cities, and real-time updates of urban traffic maps
The public 2D semantic labeling contest Potsdam dataset and Vaihingen dataset are elaborately labeled, which are provided by the International Society for Photogrammetry and Remote Sensing (ISPRS) Commission II/4
The two datasets consist of four-band image data and corresponding digital surface model (DSM) data
Summary
Building detection plays an important role in urban development planning, urban infrastructure planning, urban land use and management, land use change monitoring, digital cities, and real-time updates of urban traffic maps. Traditional algorithms for building detection based on remote sensing images are mainly driven by visual features via bottom-up approaches. These methods, such as geometric boundary-based [1], image segmentation-based [2], and building-specific auxiliary information (shadows, elevations, etc.) based [3], consider a building as a combination of low-level features that merge a building as a whole under some rules. These methods focus on the characteristic feature of buildings. The feature design requires experimental testing in decision making, thereby increasing the algorithm complexity
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
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.