Search results for "Point Cloud"

showing 10 items of 81 documents

CNN-based People Detection in Voxel Space using Intensity Measurements and Point Cluster Flattening

2021

In this paper real-time people detection is demonstrated in a relatively large indoor industrial robot cell as well as in an outdoor environment. Six depth sensors mounted at the ceiling are used to generate a merged point cloud of the cell. The merged point cloud is segmented into clusters and flattened into gray-scale 2D images in the xy and xz planes. These images are then used as input to a classifier based on convolutional neural networks (CNNs). The final output is the 3D position (x,y,z) and bounding box representing the human. The system is able to detect and track multiple humans in real-time, both indoors and outdoors. The positional accuracy of the proposed method has been verifi…

Physicsbusiness.industryPoint clusterComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONconvolutional neural networkQA75.5-76.95Space (mathematics)computer.software_genrehuman detectionFlatteningComputer Science ApplicationsIntensity (physics)flatteningControl and Systems EngineeringVoxelModeling and SimulationElectronic computers. Computer sciencepoint cloudsComputer visionArtificial intelligencebusinesscomputerSoftwareModeling, Identification and Control
researchProduct

Replication Data for: Automatic Calibration of an Industrial RGB-D Camera Network using Retroreflective Fiducial Markers.

2019

Replication Data in the form of a Robot Operating System (ROS) recording (ROS-bag) to replicate the results of the paper "Automatic Calibration of an Industrial RGB-D Camera Network using Retroreflective Fiducial Markers." The contents of the dataset are timestamped images and point clouds recorded from six different sensor nodes.

Point cloudComputingMethodologies_PATTERNRECOGNITIONEngineeringRegistrationComputer and Information ScienceVisionCalibrationComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONRGB-D
researchProduct

Novel algorithms for 3D surface point cloud boundary detection and edge reconstruction

2019

Abstract Tessellated surfaces generated from point clouds typically show inaccurate and jagged boundaries. This can lead to tolerance errors and problems such as machine judder if the model is used for ongoing manufacturing applications. This paper introduces a novel boundary point detection algorithm and spatial FFT-based filtering approach, which together allow for direct generation of low noise tessellated surfaces from point cloud data, which are not based on pre-defined threshold values. Existing detection techniques are optimized to detect points belonging to sharp edges and creases. The new algorithm is targeted at the detection of boundary points and it is able to do this better tha…

PolynomialBoundary detection Edge reconstruction Point-cloudComputer scienceTKFast Fourier transformComputational MechanicsPoint cloudBoundary (topology)02 engineering and technologySettore ING-IND/14 - Progettazione Meccanica E Costruzione Di Macchine0203 mechanical engineeringlcsh:TA1740202 electrical engineering electronic engineering information engineeringEngineering (miscellaneous)Function (mathematics)lcsh:Engineering designComputer Graphics and Computer-Aided DesignHuman-Computer InteractionComputational MathematicsNoise020303 mechanical engineering & transportsModeling and SimulationCurve fittingArtificial noise020201 artificial intelligence & image processingAlgorithmJournal of Computational Design and Engineering
researchProduct

Assessment of Classifiers and Remote Sensing Features of Hyperspectral Imagery and Stereo-Photogrammetric Point Clouds for Recognition of Tree Specie…

2018

Recognition of tree species and geospatial information on tree species composition is essential for forest management. In this study, tree species recognition was examined using hyperspectral imagery from visible to near-infrared (VNIR) and short-wave infrared (SWIR) camera sensors in combination with a 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum with a diverse selection of 26 tree species from 14 genera was used as a test area. Aerial hyperspectral imagery and high spatial resolution photogrammetric color imagery were acquired from the test area using unmanned aerial vehicle (UAV) borne sensors. Hyperspectral imagery was processed to calibrated …

Reflectance calibration010504 meteorology & atmospheric sciencesInfraredComputer sciencegeneettiset algoritmitUAVta1171Point clouddense point cloud01 natural scienceshyperspectral imagery; tree species recognition; photogrammetry; dense point cloud; reflectance calibration; UAV; random forest; genetic algorithm; machine learningilmakuvakartoitusMachine learninggenetic algorithmImage sensorfotogrammetria0105 earth and related environmental sciencesRemote sensingta113040101 forestryta213tree species recognitionspektrikuvausSpecies diversityHyperspectral imaging04 agricultural and veterinary sciencesOtaNanoreflectance calibrationDense point cloudVNIRRandom forestTree (data structure)hyperspectral imagerykoneoppiminenPhotogrammetryGenetic algorithmHyperspectral imageryPhotogrammetryTree species recognitionlajinmääritys0401 agriculture forestry and fisheriesGeneral Earth and Planetary SciencesRGB color modelkaukokartoituspuustorandom forestRandom forestRemote Sensing; Volume 10; Issue 5; Pages: 714
researchProduct

3D Reconstruction of Dynamic Vehicles using Sparse 3D-Laser-Scanner and 2D Image Fusion

2016

International audience; Map building becomes one of the most interesting research topic in computer vision field nowadays. To acquire accurate large 3D scene reconstructions, 3D laser scanners are recently developed and widely used. They produce accurate but sparse 3D point clouds of the environments. However, 3D reconstruction of rigidly moving objects along side with the large-scale 3D scene reconstruction is still lack of interest in many researches. To achieve a detailed object-level 3D reconstruction, a single scan of point cloud is insufficient due to their sparsity. For example, traditional Iterative Closest Point (ICP) registration technique or its variances are not accurate and rob…

RegistrationComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONPoint cloud02 engineering and technologyIterative reconstructionRANSAC[ INFO.INFO-CV ] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV][INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Robustness (computer science)Point Cloud0202 electrical engineering electronic engineering information engineeringComputer visionImage fusionbusiness.industry3D reconstruction[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Iterative closest point2D camera020207 software engineeringICP3D cameraMaxima and minimaGeography020201 artificial intelligence & image processingArtificial intelligencebusiness3D Reconstruction
researchProduct

Process parameters influence in additive manufacturing

2016

Additive manufacturing is a rapidly expanding technology. It allows the creation of very complex 3D objects by adding layers of material, in spite of the traditional production systems based on the removal of material. The development of additive technology has produced initially a generation of additive manufacturing techniques restricted to industrial applications, but their extraordinary degree of innovation has allowed the spreading of household systems. Nowadays, the most common domestic systems produce 3D parts through a fused deposition modeling process. Such systems have low productivity and make, usually, objects with no high accuracy and with unreliable mechanical properties. Thes…

Reverse engineering0209 industrial biotechnologyComputer scienceAdditive manufacturingmedia_common.quotation_subjectPoint cloud3D printingCAD02 engineering and technologycomputer.software_genrelaw.inventionSet (abstract data type)020901 industrial engineering & automation0203 mechanical engineeringlawQuality (business)Process engineeringReverse engineeringmedia_commonFused deposition modelingbusiness.industryProcess parameterProcess (computing)3D printing020303 mechanical engineering & transportsbusinesscomputer
researchProduct

Computer vision-based approach for rite decryption in old societies

2015

International audience; This paper presents an approach to determine the spatial arrangement of bones of horses in an excavation site and perform the 3D reconstruction of the scene. The relative 3D positioning of the bones was computed exploiting the information in images acquired at different levels, and used to relocate provided 3D models of the bones. A novel semi-supervised approach was proposed to generate dense point clouds of the bones from sparse features. The point clouds were later matched with the given models using Iterative Closest Point (ICP).

RiteComputer sciencebusiness.industry[INFO.INFO-RB] Computer Science [cs]/Robotics [cs.RO]3D reconstructionFeature extraction[ INFO.INFO-RB ] Computer Science [cs]/Robotics [cs.RO]Point cloudComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONIterative closest pointExcavationIterative reconstructionSolid modeling[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO]Computer visionArtificial intelligencebusinessComputingMethodologies_COMPUTERGRAPHICS
researchProduct

Relationship between resolution and accuracy of four intraoral scanners in complete-arch impressions

2018

Background The scanner does not measure the dental surface continually. Instead, it generates a point cloud, and these points are then joined to form the scanned object. This approximation will depend on the number of points generated (resolution), which can lead to low accuracy (trueness and precision) when fewer points are obtained. The purpose of this study is to determine the resolution of four intraoral digital imaging systems and to demonstrate the relationship between accuracy and resolution of the intraoral scanner in impressions of a complete dental arch. Material and methods A master cast of the complete maxillary arch was prepared with different dental preparations. Using four di…

ScannerComputer sciencePoint cloud02 engineering and technology03 medical and health sciencessymbols.namesake0302 clinical medicineSoftware0202 electrical engineering electronic engineering information engineeringmedicineComputer visionGeneral DentistryProsthetic Dentistrybusiness.industryResearchResolution (electron density)Digital imaging030206 dentistry:CIENCIAS MÉDICAS [UNESCO]Pearson product-moment correlation coefficientImpressionDental archmedicine.anatomical_structureUNESCO::CIENCIAS MÉDICASsymbols020201 artificial intelligence & image processingArtificial intelligencebusiness
researchProduct

3D Reconstruction of rough terrain for USARSim using a height-map method

2008

In this paper, a process for a simplified reconstruction of rough terrains from point clouds acquired using laser scanners is presented. The main idea of this work is to build height-maps which are level gray-scale images representing the ground elevation. These height-maps are generated from step-fields which can be represented by a set of side-by-side pillars. Although height-maps are a practical means for rough terrain reconstruction, it is not possible to represent two different elevations for a given location with one height-map. This is an important drawback as terrain point clouds can show different zones representing surfaces above other surfaces.In this paper, a methodology to crea…

Set (abstract data type)Computer sciencebusiness.industry3D reconstructionProcess (computing)Point cloudElevationRobotComputer visionTerrainArtificial intelligencebusinessAutomationProceedings of the 8th Workshop on Performance Metrics for Intelligent Systems
researchProduct

Detection and Isolation of Switches in Point Clouds of the German Railway Network

2015

In order to obtain an automated system of railway management, it is necessary to automatically detect, isolate and identify all switches in a point cloud which represents the railway. To realize this automated system of detection, a set of pre-processing steps is applied. The system begins by detecting and isolating tracks through application of a mask on each section of the point cloud. Then, it does a denoising through mathematical morphology and a compression in replacing a group of points by their centroid. Finally, it closes tracks holes through extrapolation. After that, the system does a low-level processing to search for all intersections between tracks, and records information on t…

Set (abstract data type)business.industryComputer sciencePoint cloudCentroidSegmentationComputer visionArtificial intelligenceIsolation (database systems)Image segmentationMathematical morphologybusinessElectronic mail2015 11th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)
researchProduct