In this paper, we propose a mesh-free numerical method for solving elliptic PDEs on unknown manifolds, identified with randomly sampled point cloud data. The PDE solver is formulated as a spectral ...method where the test function space is the span of the leading eigenfunctions of the Laplacian operator, which are approximated from the point cloud data. While the framework is flexible for any test functional space, we will consider the eigensolutions of a weighted Laplacian obtained from a symmetric Radial Basis Function (RBF) method induced by a weak approximation of a weighted Laplacian on an appropriate Hilbert space. In this paper, we consider a test function space that encodes the geometry of the data yet does not require us to identify and use the sampling density of the point cloud. To attain a more accurate approximation of the expansion coefficients, we adopt a second-order tangent space estimation method to improve the RBF interpolation accuracy in estimating the tangential derivatives. This spectral framework allows us to efficiently solve the PDE many times subjected to different parameters, which reduces the computational cost in the related inverse problem applications. In a well-posed elliptic PDE setting with randomly sampled point cloud data, we provide a theoretical analysis to demonstrate the convergence of the proposed solver as the sample size increases. We also report some numerical studies that show the convergence of the spectral solver on simple manifolds and unknown, rough surfaces. Our numerical results suggest that the proposed method is more accurate than a graph Laplacian-based solver on smooth manifolds. On rough manifolds, these two approaches are comparable. Due to the flexibility of the framework, we empirically found improved accuracies in both smoothed and unsmoothed Stanford bunny domains by blending the graph Laplacian eigensolutions and RBF interpolator.
•Solving elliptic PDEs on manifolds identified with random point cloud data.•An efficient PDE solver based on Galerkin framework.•Approximate eigenfunctions on manifold using Radial Basis Function approximation.
•Terrestrial laser scanning can be used to detect Scots pines infected by root rot.•Detection bases on structural differences interpreted from point cloud data.•Diseased individuals are characterized ...by swollen butts and deteriorated crowns.•These two symptoms are not necessarily co-existing.•Calculated structural features have potential for automatized identification.
Root rot, caused by the decay fungus Heterobasidion annosum, damages both below- and above-ground parts of Scots pines (Pinus Sylvestris L.). The diseased pines are often first characterized by deteriorated crowns and they will eventually be killed by the infection, but the process is gradual and difficult to be observed before the symptoms are severe. We tested the applicability of point cloud data produced by terrestrial laser scanning (TLS) for quantifying the structural differences between the healthy and the diseased trees. This approach was applied in a mature pine stand in southern Finland, which was known to be infected by H. annosum. We first scanned the stand using TLS, and thereafter felled the trees for detailed inspection and classification of the infection status. From the TLS point cloud, we estimated i) crosscut areas within the lowest 1 m of the stem, identifying potential deformations initiated by the fungus, ii) degree of crown deterioration, often providing the first visual signs of the infection at the level of individual trees, and iii) crown occupancy and open space around the trees, prone to be altered by the mycelial spread of the fungus between the adjacent trees. The results indicate that differences in both stem dimensions and crown deterioration can be detected between the healthy and the diseased trees. The diseased trees were found to have a more swollen butt, but no irregularities in circularity of the crosscuts were detected. In terms of vertical point distribution, the diseased trees had point accumulations at substantially greater heights, reflecting easier penetration of laser beams and sparsity of the crown. Regarding to crown occupancy, the diseased trees had more open space around their crowns, but difference to the healthy trees was not statistically significant. According to a simple prediction test based on the calculated features, up to 85% classification accuracy of the infection status was reached. This study is the first indication that TLS can successfully be applied for detecting structural changes of Scots pines connected to Heterobasidion root rot. Our results also show evidence that H. annosum causes butt swelling, which has rarely been reported as a symptom for Scots pines.
The intelligent recognition of ship geometric features is a prerequisite for enabling computers to automatically generate and deform ship hull surfaces according to requirements, thereby replacing ...the work of human designers to improve design efficiency. This paper aims to research the recognition of geometric features in three-dimensional ship data using PointNet. To achieve this goal, we first construct two ship point cloud datasets suitable for global feature classification and feature part segmentation of three-dimensional hulls. Subsequently, we conducted recognition capability testing to determine the optimal hyperparameters for identifying ship feature networks. Finally, we employ ship models with non-standard positions to implement data augmentation, enhancing the network's robustness in recognizing the initial positions of ships and achieving rapid cognition of three-dimensional ship geometric features. The findings of this research will provide technical support for ship design based on artificial intelligence technology.
Purpose
Virtual reality (VR) can be useful in explaining diseases and complications that affect children in order to improve medical communications with this vulnerable patient group. So far, ...children and young people’s responses to high‐end medical VR environments have never been assessed.
Methods
An unprecedented number of 320 children and young people were given the opportunity to interact with a VR application displaying original ophthalmic volume data via a commercially available tethered head‐mounted display (HMD). Participants completed three surveys: demographics and experience with VR, usability and perceived utility of this technology and the Simulator Sickness Questionnaire. The second survey also probed participants for suggestions on improvements and whether this system could be useful for increasing engagement in science.
Results
A total of 206 sets of surveys were received. 165 children and young people (84 female) aged 12–18 years (mean, 15 years) completed surveys that could be used for analysis. 69 participants (47.59%) were VR‐naïve, and 76 (52.41%) reported that they had previous VR experience. Results show that VR facilitated understanding of ophthalmological complications and was reasonably tolerated. Lastly, exposure to VR raised children and young people’s awareness and interest in science.
Conclusions
The VR platform used was successfully utilized and was well accepted in children to display and interact with volume‐rendered 3D ophthalmological data. Virtual reality (VR) is suitable as a novel image display platform in ophthalmology to engage children and young people.
•A color based region growing segmentation method was applied to extract the independent point cloud of apple, branches and leaves from the scene.•We proposed an improved 3D descriptor which is ...composed by color feature and 3D geometric feature to describe apples, branches and leaves from point cloud data.•We proposed an automatic recognition method based on genetic algorithm optimized SVM to recognize three classes of data.•We discussed the feasibility of using the proposed method to estimate the blocking of apples.
Accurate apple recognition is a vital step in the operation of robotic fruit picking. To improve robot recognition ability and perception in three-dimensional (3D) space, an automatic recognition method was proposed to achieve apple recognition from point cloud data. First, an improved 3D descriptor (Color-FPFH) with the fusion of color features and 3D geometry features was extracted from the preprocessed point clouds. Then, a classification category was subdivided into apple, branch, and leaf to provide the system with a more comprehensive perception capability. A classifier based on the support vector machine, optimized using a genetic algorithm, was trained by the three data classes. Finally, the results of recognition and lateral comparison were obtained by comparison with the different 3D descriptors and other classic classifiers. The results showed that the proposed method exhibited better performance. In addition, the feasibility of estimating the occurrence of blocking using proposed method was discussed.
A Graph-CNN for 3D Point Cloud Classification Zhang, Yingxue; Rabbat, Michael
2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP),
04/2018
Conference Proceeding
Open access
Graph convolutional neural networks (Graph-CNNs) extend traditional CNNs to handle data that is supported on a graph. Major challenges when working with data on graphs are that the support set (the ...vertices of the graph) do not typically have a natural ordering, and in general, the topology of the graph is not regular (i.e., vertices do not all have the same number of neighbors). Thus, Graph-CNNs have huge potential to deal with 3D point cloud data which has been obtained from sampling a manifold. In this paper we develop a Graph-CNN for classifying 3D point cloud data, called PointGCN 1 . The architecture combines localized graph convolutions with two types of graph downsampling operations (also known as pooling). By the effective exploration of the point cloud local structure using the Graph-CNN, the proposed architecture achieves competitive performance on the 3D object classification benchmark ModelNet, and our architecture is more stable than competing schemes.
Current state-of-the-art point cloud data management (PCDM) systems rely on a variety of parallel architectures and diverse data models. The main objective of these implementations is achieving ...higher scalability without compromising performance. This paper reviews the scalability and performance of state-of-the-art PCDM systems with respect to both parallel architectures and data models. More specifically, in terms of parallel architectures, shared-memory architecture, shared-disk architecture, and shared-nothing architecture are considered. In terms of data models, relational models, and novel data models (such as wide-column models) are considered. New structured query language (NewSQL) models are considered. The impacts of parallel architectures and data models are discussed with respect to theoretical perspectives and in the context of existing PCDM implementations. Based on the review, a methodical approach for the selection of parallel architectures and data models for highly scalable and performance-efficient PCDM system development is proposed. Finally, notable research gaps in the PCDM literature are presented as possible directions for future research.
Streetlights serve as fundamental infrastructure to meet the lighting needs of people on every road. However, their extensive deployment often results in unnecessary energy waste, with many ...streetlights maintaining high brightness despite minimal usage during the night. This study aims to develop a smart energy-efficient streetlight system that automatically adjusts lighting levels based on the absence of vehicles and pedestrians, detected after a 3-minute countdown. Specifically, the study utilizes mmWave radar to collect point cloud data, which undergoes denoising through Doppler, DBSCAN, and XYZ techniques. Additionally, the mmWave radar assists in training an LSTM model to identify pedestrian pathways. The implementation of the proposed system significantly reduces energy consumption and annual costs by automatically dimming or turning off streetlights in areas with minimal pedestrian activity during nighttime.
Nowadays, with the development of demand for construction, the use of composite structures in architectures become more and more popularity. How to improve the intelligent level of deformation ...monitoring has become one of the key problems. A detailed understanding about deformation behavior is significant for better monitoring of structures, especially in terms of accuracy and detail. The innovation of this paper focuses on that terrestrial laser scanning (TLS) measurement is adopted to investigate deformation of the composite masonry structures.
In this paper, deformation segmentation and analysis of the masonry structures are investigated and the deformation tendency is compared and analyzed, based on the intelligent data extraction by window selection method, where high precision 3D laser technology provides reliable experimental data for this research. The deformation of different surfaces of a composite arch is considered and the maximum displacement distribution is analyzed through partially comparing the deformation of different epoch data.
High-precision 3D point cloud data have various analyses and application use cases. This study aimed to achieve a more precise noise reduction of the raw 3D point cloud data of asphalt pavements ...obtained using 3D laser scanning. Hence, a noise-reduction algorithm integrating improved Gaussian filtering and coefficient of variation was developed. A portable laser scanner was used to collect raw, high-precision 3D point cloud data of surface textures from pavement slab samples prepared with three different types of asphalt mixtures: AC-13, SMA-13, and OGFC-13, as well as asphalt from the test sections of the Yakang Expressway. An improved Gaussian filtering and Gaussian filtering that extracts noise using the coefficient of variation were used to filter out the obvious outlier noise and small-scale burr noise, respectively. Finally, the filtering effect of the proposed algorithm, Gaussian filtering, median filtering, and mean filtering on raw 3D point cloud data of pavement textures was evaluated through subjective visual quality and objective index evaluations. The results showed that the proposed algorithm filters out noise while preserving the micro-texture structure information, outperforming Gaussian filtering, median filtering, and mean filtering.