Geometric preprocessing of noisy point sets: an experimental study

Vanco, Marek; Brunnett, Guido

doi:10.1007/s00607-006-0212-0

Cited by 7 publications

(13 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Except for the boundary corner and saddle cases, the dot product of the expected and found normal is > 0.95. We compare our results to three approaches: plane-fitting normal reconstruction (Mitra and Nguyen, 2003) and quadratic and cubic fitting (Vančo and Brunnett, 2007). These three approaches perform well on flat regions, but deteriorate as the curvature increases, as the plots show.…”

Section: Known Data Setsmentioning

confidence: 93%

“…This approach can cause problems when the surface "folds back" on itself because nearby points in Euclidean space may not be close from a geodesic measure. This problem is exacerbated by large k. One solution to this is to use an approximation of geodesic distance to build large neighborhoods from small ones (Vančo and Brunnett, 2007). We have experimented with this approach but have found it to decrease the quality and accuracy of the normal reconstruction because a smaller neighborhood reduces the chance of "bridging" a gap in the sampling and increases the effect of noise.…”

Section: Related Workmentioning

confidence: 99%

“…We categorize the normals based on feature shape in order to more clearly show that the performance depends on the local shape. We compare with the (local) SVD (Mitra and Nguyen, 2003) reconstruction (Vančo and Brunnett, 2007) and with the (global) Delaunay approach . The SVD and quadratic fits were calculated using one round of normal-based weighting (Vančo and Brunnett, 2007).…”

Section: Real-data Comparisonmentioning

confidence: 99%

“…We compare with the (local) SVD (Mitra and Nguyen, 2003) reconstruction (Vančo and Brunnett, 2007) and with the (global) Delaunay approach . The SVD and quadratic fits were calculated using one round of normal-based weighting (Vančo and Brunnett, 2007). The sampling of the dragon is much denser than the bunny, with a corresponding increase in accuracy for all algorithms.…”

Section: Real-data Comparisonmentioning

confidence: 99%

“…They then calculate the normal by plane fitting and show that adaptively choosing k increases the accuracy of the normal estimation. The fitting approach was extended to quadratic and cubic surfaces with normal-based weights (Vančo and Brunnett, 2007) which provide a better approximation in the presence of noise. A recent survey (Klasing et al, 2009) also found quadratic fitting more accurate for moderate noise, although for very high noise planar fitting was better.…”

Section: Related Workmentioning

confidence: 99%

See 4 more Smart Citations

Shape classification and normal estimation for non-uniformly sampled, noisy point data

Grimm

Smart

2011

Computers & Graphics

View full text Add to dashboard Cite

Section: Known Data Setsmentioning

confidence: 93%

Section: Related Workmentioning

confidence: 99%

Section: Real-data Comparisonmentioning

confidence: 99%

Section: Real-data Comparisonmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

Shape classification and normal estimation for non-uniformly sampled, noisy point data

Grimm

Smart

2011

Computers & Graphics

View full text Add to dashboard Cite

Distance field computation for geological slab surface data sets

Vanco

Hamann

Kreylos

et al. 2011

Comput. Visual Sci.

View full text Add to dashboard Cite

The three-dimensional shapes of tectonic plates that sink into the Earth's mantle (slabs) are the starting point for a range of geoscience studies, from determining the forces driving the motion of tectonic plates, to potential seismic and tsunami hazards, to the sources of magmas beneath active volcanos. For many of these applications finite element methods are used to model the deformation or fluid flow, and therefore the input model parameters, such as feature geometries, temperature or viscosity, must be defined with respect to a smooth, continuous distance field around the slab. In this paper we present a framework for processing sparse and noisy seismic data (earthquake locations), defining the shape of the slab and computing a continuous distance function on a mesh with variable node spacing. Due to the inhomogeneous volumetric distribution of earthquakes within the slab and significant inaccuracies in the locations of earthquakes occurring hundreds of kilometers below the Earth's surface, the seismicity data set is extremely noisy and incomplete. Therefore, the preprocessing is the major part of the framework consisting of several steps including a point based smoothing procedure, a powerful method to use other observational constraints on slab location (e.g., seismic tomography or geologic history) to extend of the slab shape beyond earthquake data set and continuous resampling using moving least squares method. For the preprocessed point data Communicated by Gabriel Wittum.we introduce approaches for finding the three-dimensional boundary of the slab and a subdivision of the slab into quadric implicit polynomials. The resulting distance field is then compiled from distances to the piecewise continuous approximation of the slab and distances to slab boundary.

show abstract

Damage Detection on Structures Using Texture Mapped Laser Point Clouds

Guldur

Hajjar

2014

Structures Congress 2014

View full text Add to dashboard Cite

Damage due to age or accumulated damage from hazards on existing structures poses a worldwide problem. In order to evaluate the current status of aging, deteriorating and damaged structures, it is vital to accurately assess the present conditions. It is possible to capture the in situ condition of structures by using laser scanners that create dense 3D point clouds. However, extracting meaningful information from point clouds automatically is challenging, and the current state-of-the-art requires a certain amount of user-interaction. Previously in our research, widely accepted point cloud processing steps were used to divide laser scanner data into meaningful object clusters, on which several damage detection methods were applied. Our current focus is on combining useful information extracted from laser scanner data with color information since novel laser scanners are capable of recording color data mapped on 3D point clouds. The color data provides additional information in the 4 th dimension, which makes it possible to detect damage types such as cracks, corrosion, and related surface defects that are generally difficult to detect by using only laser scanner data, and also helps to track volumetric changes on structures such as spalling. Even though crack and corrosion detection from images with varying resolution are extensively researched topics, damage detection by using laser scanners with and without images is a new research area that holds many opportunities for enhancing current practice in structural health monitoring. The aim of our research is to benefit from the best features of both laser scans and images, and combine them to create an automatic and effective damage detection method, which will complement the need for skilled labor during visual inspections of damaged structures and allow documenting related information automatically. This work is of interest to researchers who are focused on structural health monitoring and interested in the use of new laser scanners and damage detection algorithms for structures, as well as practitioners, who will benefit exploring new methods for improving their understanding of laser scanning technology and its capabilities for robust damage detection and documentation.

show abstract

Geometric preprocessing of noisy point sets: an experimental study

Cited by 7 publications

References 14 publications

Shape classification and normal estimation for non-uniformly sampled, noisy point data

Shape classification and normal estimation for non-uniformly sampled, noisy point data

Distance field computation for geological slab surface data sets

Damage Detection on Structures Using Texture Mapped Laser Point Clouds

Contact Info

Product

Resources

About