As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
In the era of big data, 3D electronic maps and GIS have important applications in fields such as geographic information analysis and urban computing. However, the construction cost of electronic maps is high, especially the 3D modeling of buildings as an important feature image, which leads to a high economic threshold for the acquisition of high-precision 3D building electronics. At the same time, with the increase of map area, the efficiency of 3D reconstruction of buildings is difficult to meet the demand. In order to solve the problem of high cost and low efficiency of 3D reconstruction of buildings in GIS, this paper proposes a method for fast and low-cost 3D modeling of buildings using building image data provided by online electronic maps, using deep neural network target recognition and instance segmentation techniques to segment building facades and calculate building pixel heights from 3D building image data. Secondly, a mapping model of building pixel height to metric height is built using a simple design warp network to calculate building metric height. Finally, a method is proposed to calculate the height of the blocked buildings. The evaluation of the accuracy of the experimental results compared with the baseline data shows that the 3D building data obtained using the GIS building 3D modeling scheme studied in this paper meets the expected requirements in terms of the location and shape of the bottom edge contour of the building and the height of the building, and can meet the demand for 3D building electronic map data in areas such as urban planning and wireless network optimization.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.