KEYWORDS: 3D modeling, Buildings, Data modeling, 3D scanning, Feature extraction, Error analysis, Laser scanners, 3D metrology, Principal component analysis
With the high accuracy and reliability, laser scanning has emerged as a powerful platform for surveying, providing surveyors a faster and more convenient measure method. The resulting 3D point cloud could describe the geometric characteristics of the target object in detail, which can well overcome the shortcomings of traditional survey production. This paper proposes a scheme for the geometric reconstruction of buildings based on ground-based 3D laser scanning technology, whose overall process includes: (1) Point cloud registration. (2) Calculation of dimension feature and extraction of building contour lines (3) Build the 3D model by importing the contour lines into SketchUp. (4) Modeling accuracy evaluation. The experimental results show that the method of extracting contour lines through the combination of dimension features and Otsu method is feasible, and the accuracy of the model built by this scheme is reliable.
Terrestrial laser scanning (TLS) data collected in urban scenes contain variable point densities, bringing great challenges for automatic extraction of urban objects. Aiming at the problem of the density variation of large-scale TLS data, this paper improves density-adaptive feature, and the number of projection points is replaced by relative density feature. The vertical slice feature is added into our method to form a 20-dimensional feature vector together with traditional 3D features and 2D features. Random Forest is used to identify the class of each point and evaluate the importance of each feature type. The experimental results show that the feature extraction algorithm in this paper can perform semantic segmentation on large-scale TLS data in urban scene more accurately than commonly used feature combinations, especially in man-made terrain, low vegetation, buildings, hard scape and scanning artifact classes, with an overall accuracy of 97.35%.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.