Deep Learning-Based Generation of Building Stock Data from Remote Sensing for Urban Heat Demand Modeling

Abstract

Cities are responsible for a large share of the global energy consumption. A third of the total greenhouse gas emissions are related to the buildings sector, making it an important target for reducing urban energy consumption. Detailed data on the building stock, including the thermal characteristics of individual buildings, such as the construction type, construction period, and building geometries, can strongly support decision-making for local authorities to help them spatially localize buildings with high potential for thermal renovations. In this paper, we present a workflow for deep learning-based building stock modeling using aerial images at a city scale for heat demand modeling. The extracted buildings are used for bottom-up modeling of the residential building heat demand based on construction type and construction period. The results for DL-building extraction exhibit F1-accuracies of 87%, and construction types yield an overall accuracy of 96%. The modeled heat demands display a high level of agreement of R2 0.82 compared with reference data. Finally, we analyze various refurbishment scenarios for construction periods and construction types, e.g., revealing that the targeted thermal renovation of multi-family houses constructed between the 1950s and 1970s accounts for about 47% of the total heat demand in a realistic refurbishment scenario.

Spatial parameters for transportation: A multi-modal approach for modelling the urban spatial structure using deep learning and remote sensing

“…They have been found to be very precise in building extraction tasks (Hui, Du, Ye, Qin, & Sui, 2019) using semantic segmentation and transfer learning (Wurm, Stark, Zhu, Weigand, & Taubenböck, 2019b). With these 2D footprints it is possible to further add height information and create 3D buildings by using an nDSM derived from satellite data (Wurm et al, 2021). A DSM contains the height of all objects on the earth's surface, both natural and artificial features are depicted.…”

HA-MPPNet: Height Aware-Multi Path Parallel Network for High Spatial Resolution Remote Sensing Image Semantic Seg-Mentation

“…For semantic segmentation of remote sensing images, 3D geometric information can be used to enhance segmentation performance. Some works have shown how the estimated DSM data providing useful additional information for building detection [19] or semantic segmentation [20]. For example, V-FuseNet [21] uses the DSM image as additional input for the segmentation task, which simultaneously learns RGB and height features through a two-stream network and then fuses the learned features from the two encode networks for final prediction.…”

“…At the same time, the self-attention-based methods [17,18] calculate a pixel-wise similarity map to capture long-range global context. Using these methods, the network can only learn 2D context appearance features, but for remote sensing images with more complex scenes, 3D geometric information is also essential [19,20]. Geometric In recent years, convolutional neural networks (CNNs) have been successfully applied in the field of remote sensing images due to their excellent performance such as building extraction [4], object detection [5], image classification [6], and so on [7].…”

“…At the same time, the self-attention-based methods [17,18] calculate a pixel-wise similarity map to capture long-range global context. Using these methods, the network can only learn 2D context appearance features, but for remote sensing images with more complex scenes, 3D geometric information is also essential [19,20]. Geometric features can discriminate those with similar 2D appearance, but significantly different 3D features objects such as trees and low vegetation, as shown in Figure 1b.…”

Detecting Pipeline Pathways in Landsat 5 Satellite Images with Deep Learning

“…Deep learning describes a set of algorithms that allows computational models to learn data representations with multiple levels of abstraction [15]. Its applications range from the improvement of industrial processes [16,17] to the support in medical diagnostics [18] to remote sensing [19,20] and computer vision [21]. In computer vision, problems like the detection of pipeline-like objects can be solved by classifying each pixel of an image to a respective object class, a process known as semantic image segmentation [22].…”