Abstract
In this study, a deep learning (DL)-based approach is proposed for the detection and reconstruction of buildings from a single aerial image. The pre-required knowledge to reconstruct the 3D shapes of buildings, including the height data as well as the linear elements of individual roofs, is derived from the RGB image using an optimized multi-scale convolutional–deconvolutional network (MSCDN). The proposed network is composed of two feature extraction levels to first predict the coarse features, and then automatically refine them. The predicted features include the normalized digital surface models (nDSMs) and linear elements of roofs in three classes of eave, ridge, and hip lines. Then, the prismatic models of buildings are generated by analyzing the eave lines. The parametric models of individual roofs are also reconstructed using the predicted ridge and hip lines. The experiments show that, even in the presence of noises in height values, the proposed method performs well on 3D reconstruction of buildings with different shapes and complexities. The average root mean square error (RMSE) and normalized median absolute deviation (NMAD) metrics are about 3.43 m and 1.13 m, respectively for the predicted nDSM. Moreover, the quality of the extracted linear elements is about 91.31% and 83.69% for the Potsdam and Zeebrugge test data, respectively. Unlike the state-of-the-art methods, the proposed approach does not need any additional or auxiliary data and employs a single image to reconstruct the 3D models of buildings with the competitive precision of about 1.2 m and 0.8 m for the horizontal and vertical RMSEs over the Potsdam data and about 3.9 m and 2.4 m over the Zeebrugge test data.
Highlights
Due to the significant advances in remote sensing technologies, the interest in the development of automatic and robust approaches to extract accurate and up-to-date 3D geo-information of land covers from remotely sensed data is rapidly increasing
Unlike recent approaches in photogrammetry and remote sensing requiring often both ortho-photos and high-resolution digital surface models (DSMs), the proposed method utilizes the power of convolutional neural network (CNN) to extract the inherent and latent features from a single image and interpret them as 3D information for building reconstruction
There were some limitations in providing the proper training datasets, two optimized multi-scale convolutional–deconvolutional network (MSCDN) were trained for height prediction and roofline segmentation tasks
Summary
Due to the significant advances in remote sensing technologies, the interest in the development of automatic and robust approaches to extract accurate and up-to-date 3D geo-information of land covers from remotely sensed data is rapidly increasing. High-resolution point cloud data generated by LiDAR, photogrammetry, or SAR technologies are not available everywhere and generation of updated DSMs needs a considerable amount of effort, time, and cost, especially for urban areas To address these issues, a knowledge-based 3DBR approach is proposed in this paper by exploiting CNNs to extract high-level information from a single image. A knowledge-based 3DBR approach is proposed in this paper by exploiting CNNs to extract high-level information from a single image This valuable knowledge includes the location of buildings, the linear elements of building roofs, such as eave, ridge, and hip lines as well as the heights (e.g., nDSMs) of buildings, which are essential for 3DBR and reduce the complexity of reconstruction. A training dataset (https://github.com/loosgagnet/Roofline-Extraction) including linear elements of different roofs is created manually, which can be used for different applications such as 3D city modeling and CAD models
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 callDisclaimer: 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.
