Normal Vector Estimation Research Articles

Fast Extraction Algorithm of Planar Targets Based on Point Cloud Data for Monitoring the Synchronization of Bridge Jacking Displacements

Transverse synchronization of vertical displacements of all jacking-up points is an important monitoring indicator to replace bearings in assembled multigirder bridges during the jacking phase. Currently, using target paper to identify the 3D coordinates of control points reduces the complexity of monitoring operations and improves the stability of data precision. However, the existing planar target locating methods have low accuracy, inefficiency, and subjectivity, which seriously hinders the construction process of bearing replacement. Accurately obtaining the center coordinates of multiple targets from a point cloud in a short monitoring period remains a challenge. This study proposes a high-precision automated algorithm to extract target center points in low-density point clouds to quickly calculate real target center points. First, we construct a standard point cloud model of the target papers for scanning, including color and geometric features. Then, we extract the measured point cloud of the typical jacking operation phase based on the reflection intensity and size information. Next, we map the intensity values of the measured point cloud into the color channel and register the measured point cloud with its standard point cloud model using the normal vector estimation and colored ICP algorithms. Finally, we extract the center point of the measured targets. Numerical experiments and engineering test results show that the proposed method converges quickly with high precision and good robustness, which saves 91.4% of the time compared with the traditional method. In general, this research can provide effective technical support for 3D laser scanning in monitoring the operation phase of bridge jacking.

Machine learning and the James–Stein estimator

It is now 62 years since the publication of James and Stein’s seminal article on the estimation of a multivariate normal mean vector. The paper made a spectacular first impression on the statistical community through its demonstration of inadmissability of the maximum likelihood estimator. It continues to be influential, but not for the initial reasons. Empirical Bayes shrinkage estimation, now a major topic, found its early justification in the James–Stein formula. Less obvious downstream topics include Tweedie’s formula and Benjamini and Hochberg’s false discovery rate algorithm. This is a short and mainly non-technical review of the James–Stein rule and its effects on the machine learning era of statistical innovation.

Polarized Object Surface Reconstruction Algorithm Based on RU-GAN Network.

There are six possible solutions for the surface normal vectors obtained from polarization information during 3D reconstruction. To resolve the ambiguity of surface normal vectors, scholars have introduced additional information, such as shading information. However, this makes the 3D reconstruction task too burdensome. Therefore, in order to make the 3D reconstruction more generally applicable, this paper proposes a complete framework to reconstruct the surface of an object using only polarized images. To solve the ambiguity problem of surface normal vectors, a jump-compensated U-shaped generative adversarial network (RU-Gan) based on jump compensation is designed for fusing six surface normal vectors. Among them, jump compensation is proposed in the encoder and decoder parts, and the content loss function is reconstructed, among other approaches. For the problem that the reflective region of the original image will cause the estimated normal vector to deviate from the true normal vector, a specular reflection model is proposed to optimize the dataset, thus reducing the reflective region. Experiments show that the estimated normal vector obtained in this paper improves the accuracy by about 20° compared with the previous conventional work, and improves the accuracy by about 1.5° compared with the recent neural network model, which means the neural network model proposed in this paper is more suitable for the normal vector estimation task. Furthermore, the object surface reconstruction framework proposed in this paper has the characteristics of simple implementation conditions and high accuracy of reconstructed texture.

Robust pooling through the data mode

The task of learning from point cloud data is always challenging due to the often occurrence of noise and outliers in the data. Such data inaccuracies can significantly influence the performance of state-of-the-art deep learning networks and their ability to classify or segment objects. While there are some robust deep-learning approaches, they are computationally too expensive for real-time applications. This paper proposes a deep learning solution that includes novel robust pooling layers which greatly enhance network robustness and perform significantly faster than state-of-the-art approaches. The proposed pooling layers replace conventional pooling layers in networks with global pooling operations such as PointNet and DGCNN. The proposed pooling layers look for data mode/cluster using two methods, RANSAC, and histogram, as clusters are indicative of models. We tested the proposed pooling layers on several tasks such as classification, part segmentation, and points normal vector estimation. The results show excellent robustness to high levels of data corruption with less computational requirements as compared to robust state-of-the-art methods. our code can be found at https://github.com/AymanMukh/ModePooling.

An automatic measurement system for the wall thickness of corrugated plate based on laser triangulation method

Wall thickness is an important parameter used in the characterization of the surface quality of chevron corrugated plates. However, current measurement techniques such as contact mechanical or destructive modes are timing-consuming and are limited by the complex morphology of the measured object. In this study, an optical measurement system is established for experimental evaluation of the wall thickness. Its optical design and working characteristic are described based on the triangulation principle. By performing normal vector estimation, and combining with the search results about the best possible reconstructed mesh surface, the wall thickness at each measurement point can be evaluated. A series of comparison experiments with the maximum deviation 0.043 mm show that our proposed method can provide effective and automatic technical support for dimensional detection of complex components.

Single-spacecraft techniques for shock parameters estimation: A systematic approach

Spacecraft missions provide the unique opportunity to study the properties of collisionless shocks utilising in situ measurements. In the past years, several diagnostics have been developed to address key shock parameters using time series of magnetic field (and plasma) data collected by a single spacecraft crossing a shock front. A critical aspect of such diagnostics is the averaging process involved in the evaluation of upstream/downstream quantities. In this work, we discuss several of these techniques, with a particular focus on the shock obliquity (defined as the angle between the upstream magnetic field and the shock normal vector) estimation. We introduce a systematic variation of the upstream/downstream averaging windows, yielding to an ensemble of shock parameters, which is a useful tool to address the robustness of their estimation. This approach is first tested with a synthetic shock dataset compliant with the Rankine-Hugoniot jump conditions for a shock, including the presence of noise and disturbances. We then employ self-consistent, hybrid kinetic shock simulations to apply the diagnostics to virtual spacecraft crossing the shock front at various stages of its evolution, highlighting the role of shock-induced fluctuations in the parameters’ estimation. This approach has the strong advantage of retaining some important properties of collisionless shock (such as, for example, the shock front microstructure) while being able to set a known, nominal set of shock parameters. Finally, two recent observations of interplanetary shocks from the Solar Orbiter spacecraft are presented, to demonstrate the use of this systematic approach to real events of shock crossings. The approach is also tested on an interplanetary shock measured by the four spacecraft of the Magnetospheric Multiscale (MMS) mission. All the Python software developed and used for the diagnostics (SerPyShock) is made available for the public, including an example of parameter estimation for a shock wave recently observed in-situ by the Solar Orbiter spacecraft.

SharpNet: A deep learning method for normal vector estimation of point cloud with sharp features

The normal vector is a basic attribute of point clouds. Traditional estimation methods are susceptible to noise and outliers. Recently, it reported that estimation robustness can be greatly improved by introducing Deep Neural Network (DNN), but how to accurately obtain the normal vector of sharp features still needs to be further studied. This paper proposes SharpNet, a DNN framework specializing in sharp features of CAD-like models, to transform problems into feature classification by the discretization of normal vector space. In order to eliminate the discretization error, a normal vector refining method is presented, which uses the difference between the initial normal vectors to distinguish neighborhood points of different local surface patches. Finally, the normal vector can be estimated accurately from the refined neighborhood points. Experiments show that our algorithm can estimate the normal vector of sharp features of CAD-like models accurately in challenging situations, and is superior to other DNN-based methods in terms of efficiency.

High-order level set reinitialization for multiphase flow simulations based on unstructured grids

Reinitialization is essential to maintaining the accuracy of the level set method at capturing interfaces. In this paper, a high-order reinitialization method for the level set function which has previously been developed for structured grids was extended to unstructured grids. The proposed method involves constructing a stencil of an unstructured grid to define a high-order polynomial that approximates a piecewise segment of the interface. Then, the closest point method is adopted to estimate the signed distance function at grid points. The proposed method was validated by solving some benchmark problems of static cases and then the evolution of moving interfacial problems subject to prescribed velocity fields for various grid resolutions consisting of triangular elements. The accuracy of the proposed method was evaluated for computing geometric quantities such as the normal vector and curvature field. The proposed method proved to have high-order accuracy for estimating not only the signed distance function but also geometric quantities for a smooth interface. Finally, a local correction procedure was developed to apply the proposed method to level set problems involving the interface with kinks. The proposed method can accurately estimate normal vector and curvature fields including singularities both for static cases and for dynamic cases where an interface experiences topological changes. • High-order reinitialization of level set method on unstructured grid is developed. • We conduct a high-order estimation of normal vector and curvature on unstructured grid. • We propose a local correction procedure to manage kinks in level set method. • We successfully solve level set problems involving topological changes with kinks.

MP58-10 A NOVEL APPROACH FOR 3D RECONSTRUCTION OF THE PROSTATE GLAND THAT ALLOWS TUMOUR LOCATION, VOLUME ESTIMATION, AND GLEASON CHARACTERIZATION

You have accessJournal of UrologyCME1 May 2022MP58-10 A NOVEL APPROACH FOR 3D RECONSTRUCTION OF THE PROSTATE GLAND THAT ALLOWS TUMOUR LOCATION, VOLUME ESTIMATION, AND GLEASON CHARACTERIZATION Pedro Oliveira, Avaneesh Meena, Saikiran Bonthu, Nitin Singhal, Ashwin Sachdeva, Yatin Jain, and Vijay Ramani Pedro OliveiraPedro Oliveira More articles by this author , Avaneesh MeenaAvaneesh Meena More articles by this author , Saikiran BonthuSaikiran Bonthu More articles by this author , Nitin SinghalNitin Singhal More articles by this author , Ashwin SachdevaAshwin Sachdeva More articles by this author , Yatin JainYatin Jain More articles by this author , and Vijay RamaniVijay Ramani More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000002641.10AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: We propose a novel approach for identification (ID) of clinical significant PCa on mpMRI based upon retrospective comparison of in vivo mpMRI images to spatially concordant digitally-scanned post-prostatectomy H&E images. Steps: 1) localisation and Gleason grading of tumour foci; 2) reconstruction of H&E slides in 3D; 3) alignment of reconstructed histology to mpMRI images; 4) labelling of aggressive prostate cancer on mpMRI images using reconstructed 3D histology; 5) training a deep learning-based model for unsupervised segmentation of aggressive PCa foci using mpMRI images and transferred labels. Using this approach, the extent of cancer can be mapped directly onto mpMRI enabling accurate segmentation of voxels corresponding to tumour foci, including ID of mpMRI-invisible lesions using radiomic features. METHODS: The step 2 is presented herein. Whole-mount histopathological sections from totally embedded radical prostatectomy specimens, with correspondent diagnostic pre-biopsy mpMRI, were used. Apex and base tissue blocks were cut perpendicular to the axial plane, with the central portion of the gland sliced in 4 mm thick sections. H&E whole-mount slides were digitised at 40x. 3D reconstruction was performed using a novel computational strategy that includes: 1) angular alignment of individual macrodissected tissue pieces using colour ink markers; 2) scale alignment to fit the pieces on a pre-defined bounding box; 3) generation of intermediate layers between two pieces; 4) normal vector estimation; and 5) Poisson reconstruction to generate the triangular mesh. RESULTS: The volume estimate from the original prostate specimen was compared to the reconstructed volume to assess the 3D reconstruction performance with a correlation of 85%-88%. Because the base and apex portions are not discarded, we establish a high correlation between the reconstructed 3D histopathological volume and actual prostate volume. Further, this methodology allowed ID not only of independent tumour foci within the gland but also 3D reconstruction of the different Gleason patterns with accurate estimation of tumour volume and prognostic Grade Group. CONCLUSIONS: A method for reconstructing 3D prostate volumes from 2D histology images has been presented. The development of radiomic and deep learning algorithms to automatically detect prostate cancer on MRI will be aided by the accurate labelling of tumour foci on mpMRI images using our 3D reconstruction approach. Source of Funding: NHS/Aira-Matrix Project PO01 © 2022 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 207Issue Supplement 5May 2022Page: e995 Advertisement Copyright & Permissions© 2022 by American Urological Association Education and Research, Inc.MetricsAuthor Information Pedro Oliveira More articles by this author Avaneesh Meena More articles by this author Saikiran Bonthu More articles by this author Nitin Singhal More articles by this author Ashwin Sachdeva More articles by this author Yatin Jain More articles by this author Vijay Ramani More articles by this author Expand All Advertisement PDF DownloadLoading ...

Query-efficient decision-based attack via sampling distribution reshaping

With a limited query budget and only the final decision of a target model, how to find adversarial examples with low-magnitude distortion has attracted great attention among researchers. Recent solutions to this issue made use of the estimated normal vector at a boundary data point to search for adversarial examples. However, since the sampling independence between two sampling epochs, they still suffer from a prohibitively high query budget, which will get worse when the dimensionality of the attacked samples get increased. To push for further development, in this paper, we pay attention to a query-efficient method to estimate the normal vector for decision-based attack in high-dimensional space. Specifically, we propose a simple yet effective normal vector estimation framework for high-dimension decision-based attack via Sampling Distribution Reshaping, dubbed SDR. Next, SDR is incorporated into general geometric attack framework. Briefly, SDR leverages all the historically sampled noise to build a guiding vector, which will be used to reshape the next sampling distribution. Besides, we also extend SDR to different ℓp norms for p={2,∞} and deploy low-frequency constraint to enhance the performance of SDR. Compared to peer decision-based attacks, SDR can reach the competitive ℓp norms for p={2,∞}, according to extensive experimental evaluations against both defended and undefended classifiers. Since the simplicity and effectiveness of SDR, we think that reshaping the sampling distribution deserves further research in future works.

Research on Pose Estimation Method of Medicine Box Based on Secondary Key Point Matching

Aiming at the difficulty and low efficiency of pose estimation in the process of automatic drug picking, a pose estimation method combining plane segmentation and key point matching is proposed. First, a point cloud hierarchical clustering algorithm is used to segment the plane area, and the improved secondary key point matching method is used to identify and locate the pill box in each thread, which is convenient for multi-thread processing. Then the point cloud plane normal vector estimation method based on least squares is used to realize the posture reconstruction of the pill box. 600 images with viewpoint change, rotation transformation and scale change were collected for testing, the experimental results show that the pose estimation speed of the medicine box can reach 11 frames/s, the positioning deviation is less than 5 mm, and the pose estimation deviation is within 2°. So the method is efficient and can accurately identify the pose of the medicine box.

Single-Stage Adaptive Multi-Scale Point Cloud Noise Filtering Algorithm Based on Feature Information

This paper proposes a single-stage adaptive multi-scale noise filtering algorithm for point clouds, based on feature information, which aims to mitigate the fact that the current laser point cloud noise filtering algorithm has difficulty quickly completing the single-stage adaptive filtering of multi-scale noise. The feature information from each point of the point cloud is obtained based on the efficient k-dimensional (k-d) tree data structure and amended normal vector estimation methods, and the adaptive threshold is used to divide the point cloud into large-scale noise, a feature-rich region, and a flat region to reduce the computational time. The large-scale noise is removed directly, the feature-rich and flat regions are filtered via improved bilateral filtering algorithm and weighted average filtering algorithm based on grey relational analysis, respectively. Simulation results show that the proposed algorithm performs better than the state-of-art comparison algorithms. It was, thus, verified that the algorithm proposed in this paper can quickly and adaptively (i) filter out large-scale noise, (ii) smooth small-scale noise, and (iii) effectively maintain the geometric features of the point cloud. The developed algorithm provides research thought for filtering pre-processing methods applicable in 3D measurements, remote sensing, and target recognition based on point clouds.

A clustering-based simplification of massive automobile-bodies point cloud for lightweight design

Adaptive simplification for massive and large-scale automobile-bodies point cloud obtained by 3D laser-scanning has been proven to be an effective technology to conduct lightweight design. This paper introduces a point-based algorithm to simplify laser-scanning point cloud without any support of fitted surface. The intrinsic characteristic of laser-scanning data is investigated to produce a topological connectivity for adjacent points in scanlines. We explore an automatic normal-vector estimation framework through the relationship between normal-vector and its adjacent geometric elements. To retain more points in high-curvature areas and fewer points in planar regions efficiently, the local normal-vector variance is adopted to determine subdivision-decision condition. The boundary points are detected and then preserved before non-uniform subdivision. A relevant simplification system based on our algorithm is developed. Many simplification cases are implemented to validate the effectiveness of our method and demonstrate the feasibility for automobile-bodies point cloud. The comparison with other point-based methods is also performed to illustrate the superiority of our method.

Multiple Cylinder Extraction from Organized Point Clouds.

Most man-made objects are composed of a few basic geometric primitives (GPs) such as spheres, cylinders, planes, ellipsoids, or cones. Thus, the object recognition problem can be considered as one of geometric primitives extraction. Among the different geometric primitives, cylinders are the most frequently used GPs in real-world scenes. Therefore, cylinder detection and extraction are of great importance in 3D computer vision. Despite the rapid progress of cylinder detection algorithms, there are still two open problems in this area. First, a robust strategy is needed for the initial sample selection component of the cylinder extraction module. Second, detecting multiple cylinders simultaneously has not yet been investigated in depth. In this paper, a robust solution is provided to address these problems. The proposed solution is divided into three sub-modules. The first sub-module is a fast and accurate normal vector estimation algorithm from raw depth images. With the estimation method, a closed-form solution is provided for computing the normal vector at each point. The second sub-module benefits from the maximally stable extremal regions (MSER) feature detector to simultaneously detect cylinders present in the scene. Finally, the detected cylinders are extracted using the proposed cylinder extraction algorithm. Quantitative and qualitative results show that the proposed algorithm outperforms the baseline algorithms in each of the following areas: normal estimation, cylinder detection, and cylinder extraction.

Unstructured un-split geometrical Volume-of-Fluid methods – A review

Geometrical Volume-of-Fluid (VoF) methods mainly support structured meshes, and only a small number of contributions in the scientific literature report results with unstructured meshes and three spatial dimensions. Unstructured meshes are traditionally used for handling geometrically complex solution domains that are prevalent when simulating problems of industrial relevance. However, three-dimensional geometrical operations are significantly more complex than their two-dimensional counterparts, which is confirmed by the ratio of publications with three-dimensional results on unstructured meshes to publications with two-dimensional results or support for structured meshes. Additionally, unstructured meshes present challenges in serial and parallel computational efficiency, accuracy, implementation complexity, and robustness. Ongoing research is still very active, focusing on different issues: interface positioning in general polyhedra, estimation of interface normal vectors, advection accuracy, and parallel and serial computational efficiency.This survey tries to give a complete and critical overview of classical, as well as contemporary geometrical VOF methods with concise explanations of the underlying ideas and sub-algorithms, focusing primarily on unstructured meshes and three dimensional calculations. Reviewed methods are listed in historical order and compared in terms of accuracy and computational efficiency.

Robust normal vector estimation in 3D point clouds through iterative principal component analysis

This paper introduces a robust normal vector estimator for point cloud data. It can handle sharp features as well as smooth areas. Our method is based on the inclusion of a robust estimator into a Principal Component Analysis in the neighborhood of the studied point, so that it can detect and reject outliers automatically during the estimation. A projection process ensures robustness against noise. Two automatic initializations are computed, leading to independent optimizations making the algorithm robust to neighborhood anisotropy around sharp features. An evaluation has been carried out in which the algorithm is compared to state-of-the-art methods. The results show that it is more robust against low and/or non-uniform samplings, high noise levels and outliers. Moreover, our algorithm is fast relative to existing methods handling sharp features. The code is available on the website: https://projet.liris.cnrs.fr/pcr/, and integrated in the platform: https://github.com/MEPP-team/MEPP2.

Reflectance Transformation Imaging as a tool for engineering failure analysis

Reflectance Transformation Imaging (RTI) is a technique which can be used to record the surface appearance of objects. It can capture a surface's interaction with incident light from different directions and it enables estimation of the surface normal vector. 3D surface profile information can be estimated from RTI datasets using photometric stereo methods. RTI currently has applications in archaeometry and cultural heritage studies but here we discuss the use of RTI in an engineering context: as a tool for failure analysis and particularly for examining failure surfaces in structural materials and components. We demonstrate that RTI can be a practical and low-cost technique for failure analysis, complementing other surface analysis and 3D scanning methods.

Automatic Conformal Anti-radial Ultrasound Scanning for Whole Breast Screening

BackgroundAnti-radial ultrasound scanning is one of the main scanning approaches used in ultrasound breast screening. It can be used for cross-sectional imaging of mammary ductal/lobular tissue and provide information about suspicious tissue. It has certain advantages compared to linear scanning, but automatic anti-radial scanning is not yet available. Our goal is to propose an ultrasound scanning system for whole breast anti-radial scanning.MethodsWe previously developed an automatic ultrasound scanning system for whole breast screening involving linear scanning. The present study builds on our previous work by incorporating (1) surface-reconstruction algorithms, (2) a rotatable holder design, and (3) a scan-path-smoothing algorithm to achieve conformal anti-radial scanning.ResultsAn improvement of approximately 40% in the normal-vector estimation is obtained with our new method, and the scan stability is improved by the scan-path-smoothing algorithm. 3D volume data of each scan are available.ConclusionsWe have successfully developed an automatic ultrasound scanning system for anti-radial breast scanning. This type of system, which has not been reported previously in the literature, can be an effective tool for fully automatic ultrasound breast screening.

A coupled phase field framework for solving incompressible two-phase flows

In this paper a coupled phase field framework will be proposed for simulating the incompressible two-phase flows. For tracking the interface between two different phases, the improved conservative phase field equation is proposed, while the non-linear preconditioning phase field equation is utilized for alleviating the issue of erroneous estimation of normal vector. The proposed method is then incorporated with the improved divergence-free condition compensated coupled framework for solving two phase flows. With the present quasi multi moment framework, not only can the pressure and velocity be fully coupled, the balanced-forced surface tension model can also be employed without any issue. In order to show the applicability and accuracy, benchmark problems will first be investigated by the present framework, before a more comprehensive comparison with other experimental results and numerical predictions.

A Physics-Based Estimation of Mean Curvature Normal Vector for Triangulated Surfaces

In this note, we derive an approximation for the mean curvature normal vector on vertices of triangulated surface meshes from the Young-Laplace equation and the force balance principle. We then demonstrate that the approximation expression from our physics-based derivation is equivalent to the discrete Laplace-Beltrami operator approach in the literature. This work, in addition to providing an alternative expression to calculate the mean curvature normal vector, can be further extended to other mesh structures, including non-triangular and heterogeneous meshes.