A Novel Method Based on Deep Learning, GIS and Geomatics Software for Building a 3D City Model from VHR Satellite Stereo Imagery

Massimiliano Pepe, Domenica Costantino, Vincenzo Saverio Alfio, Gabriele Vozza, Elena Cartellino

ISPRS International Journal of Geo-Information · 2021 · 54 citations · 37 references

DOIFull text

Open access

Concepts

TL;DR

The study seeks a suitable method to construct a 3D city model from stereo satellite imagery. The authors develop a workflow that first pan‑sharpened WorldView‑2 images, then uses deep learning to detect building footprints, and finally applies a GIS algorithm to extract building heights from the DSM. The workflow yields high‑quality color images with Gram–Schmidt pan‑sharpening, accurately identifies footprints (precision, recall, overall accuracy, F1), and produces fast, precise 3D city models of prismatic buildings with flat roofs.

Abstract

The aim of the paper is to identify a suitable method for the construction of a 3D city model from stereo satellite imagery. In order to reach this goal, it is necessary to build a workflow consisting of three main steps: (1) Increasing the geometric resolution of the color images through the use of pan-sharpening techniques, (2) identification of the buildings’ footprint through deep-learning techniques and, finally, (3) building an algorithm in GIS (Geographic Information System) for the extraction of the elevation of buildings. The developed method was applied to stereo imagery acquired by WorldView-2 (WV-2), a commercial Earth-observation satellite. The comparison of the different pan-sharpening techniques showed that the Gram–Schmidt method provided better-quality color images than the other techniques examined; this result was deduced from both the visual analysis of the orthophotos and the analysis of quality indices (RMSE, RASE and ERGAS). Subsequently, a deep-learning technique was applied for pan sharpening an image in order to extract the footprint of buildings. Performance indices (precision, recall, overall accuracy and the F1measure) showed an elevated accuracy in automatic recognition of the buildings. Finally, starting from the Digital Surface Model (DSM) generated by satellite imagery, an algorithm built in the GIS environment allowed the extraction of the building height from the elevation model. In this way, it was possible to build a 3D city model where the buildings are represented as prismatic solids with flat roofs, in a fast and precise way.

References

37