scholarly journals CONTEXTUAL CLASSIFICATION OF POINT CLOUD DATA BY EXPLOITING INDIVIDUAL 3D NEIGBOURHOODS

2015 ◽  
Vol II-3/W4 ◽  
pp. 271-278 ◽  
Author(s):  
M. Weinmann ◽  
A. Schmidt ◽  
C. Mallet ◽  
S. Hinz ◽  
F. Rottensteiner ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Point Cloud ◽  
Contextual Information ◽  
Point Clouds ◽  
Optimal Size ◽  
Point Cloud Data ◽  
Cloud Data ◽  
Semantic Class ◽  
3D Point Clouds

The fully automated analysis of 3D point clouds is of great importance in photogrammetry, remote sensing and computer vision. For reliably extracting objects such as buildings, road inventory or vegetation, many approaches rely on the results of a point cloud classification, where each 3D point is assigned a respective semantic class label. Such an assignment, in turn, typically involves statistical methods for feature extraction and machine learning. Whereas the different components in the processing workflow have extensively, but separately been investigated in recent years, the respective connection by sharing the results of crucial tasks across all components has not yet been addressed. This connection not only encapsulates the interrelated issues of neighborhood selection and feature extraction, but also the issue of how to involve spatial context in the classification step. In this paper, we present a novel and generic approach for 3D scene analysis which relies on (<i>i</i>) individually optimized 3D neighborhoods for (<i>ii</i>) the extraction of distinctive geometric features and (<i>iii</i>) the contextual classification of point cloud data. For a labeled benchmark dataset, we demonstrate the beneficial impact of involving contextual information in the classification process and that using individual 3D neighborhoods of optimal size significantly increases the quality of the results for both pointwise and contextual classification.

scholarly journals 3D Point Cloud Data in Conveying Information for Local Green Factor Assessment

2021 ◽  
Vol 10 (11) ◽  
pp. 762
Author(s):  
Kaisa Jaalama ◽  
Heikki Kauhanen ◽  
Aino Keitaanniemi ◽  
Toni Rantanen ◽  
Juho-Pekka Virtanen ◽  
...  
Keyword(s):  
Green Infrastructure ◽  
Point Cloud ◽  
Point Clouds ◽  
Urban Ecosystem ◽  
Small Scale ◽  
Point Cloud Data ◽  
Multidisciplinary Research ◽  
Cloud Data ◽  
3D Point Clouds ◽  
Geometrical Form

The importance of ensuring the adequacy of urban ecosystem services and green infrastructure has been widely highlighted in multidisciplinary research. Meanwhile, the consolidation of cities has been a dominant trend in urban development and has led to the development and implementation of the green factor tool in cities such as Berlin, Melbourne, and Helsinki. In this study, elements of the green factor tool were monitored with laser-scanned and photogrammetrically derived point cloud datasets encompassing a yard in Espoo, Finland. The results show that with the support of 3D point clouds, it is possible to support the monitoring of the local green infrastructure, including elements of smaller size in green areas and yards. However, point clouds generated by distinct means have differing abilities in conveying information on green elements, and canopy covers, for example, might hinder these abilities. Additionally, some green factor elements are more promising for 3D measurement-based monitoring than others, such as those with clear geometrical form. The results encourage the involvement of 3D measuring technologies for monitoring local urban green infrastructure (UGI), also of small scale.

6-DOF Localization for a Mobile Robot Using Outdoor 3D Point Clouds

2010 ◽  
Vol 22 (2) ◽  
pp. 158-166 ◽  
Author(s):  
Taro Suzuki ◽  
◽  
Yoshiharu Amano ◽  
Takumi Hashizume
Keyword(s):  
Mobile Robot ◽  
Point Cloud ◽  
Field Experiments ◽  
Point Clouds ◽  
Measurement Model ◽  
3D Point Cloud ◽  
Point Cloud Data ◽  
Cloud Data ◽  
Continuous Contact ◽  
3D Point Clouds

This paper describes outdoor localization for a mobile robot using a laser scanner and three-dimensional (3D) point cloud data. A Mobile Mapping System (MMS) measures outdoor 3D point clouds easily and precisely. The full six-dimensional state of a mobile robot is estimated combining dead reckoning and 3D point cloud data. Two-dimensional (2D) position and orientation are extended to 3D using 3D point clouds assuming that the mobile robot remains in continuous contact with the road surface. Our approach applies a particle filter to correct position error in the laser measurement model in 3D point cloud space. Field experiments were conducted to evaluate the accuracy of our proposal. As the result of the experiment, it was confirmed that a localization precision of 0.2 m (RMS) is possible using our proposal.

scholarly journals Review: Deep Learning on 3D Point Clouds

2020 ◽  
Vol 12 (11) ◽  
pp. 1729 ◽  
Author(s):  
Saifullahi Aminu Bello ◽  
Shangshu Yu ◽  
Cheng Wang ◽  
Jibril Muhmmad Adam ◽  
Jonathan Li
Keyword(s):  
Deep Learning ◽  
Point Cloud ◽  
General Structure ◽  
Point Clouds ◽  
Autonomous Driving ◽  
Point Cloud Data ◽  
3D Vision ◽  
Cloud Data ◽  
3D Point Clouds ◽  
Learning Techniques

A point cloud is a set of points defined in a 3D metric space. Point clouds have become one of the most significant data formats for 3D representation and are gaining increased popularity as a result of the increased availability of acquisition devices, as well as seeing increased application in areas such as robotics, autonomous driving, and augmented and virtual reality. Deep learning is now the most powerful tool for data processing in computer vision and is becoming the most preferred technique for tasks such as classification, segmentation, and detection. While deep learning techniques are mainly applied to data with a structured grid, the point cloud, on the other hand, is unstructured. The unstructuredness of point clouds makes the use of deep learning for its direct processing very challenging. This paper contains a review of the recent state-of-the-art deep learning techniques, mainly focusing on raw point cloud data. The initial work on deep learning directly with raw point cloud data did not model local regions; therefore, subsequent approaches model local regions through sampling and grouping. More recently, several approaches have been proposed that not only model the local regions but also explore the correlation between points in the local regions. From the survey, we conclude that approaches that model local regions and take into account the correlation between points in the local regions perform better. Contrary to existing reviews, this paper provides a general structure for learning with raw point clouds, and various methods were compared based on the general structure. This work also introduces the popular 3D point cloud benchmark datasets and discusses the application of deep learning in popular 3D vision tasks, including classification, segmentation, and detection.

An efficient simplification method for point cloud based on salient regions detection

2019 ◽  
Vol 53 (2) ◽  
pp. 487-504 ◽  
Author(s):  
Abdul Rahman El Sayed ◽  
Abdallah El Chakik ◽  
Hassan Alabboud ◽  
Adnan Yassine
Keyword(s):  
Point Cloud ◽  
Processing Time ◽  
Point Clouds ◽  
The Other ◽  
Point Cloud Data ◽  
Weighted Graphs ◽  
Cloud Data ◽  
3D Point Clouds ◽  
Salient Regions ◽  
The Stability

Many computer vision approaches for point clouds processing consider 3D simplification as an important preprocessing phase. On the other hand, the big amount of point cloud data that describe a 3D object require excessively a large storage and long processing time. In this paper, we present an efficient simplification method for 3D point clouds using weighted graphs representation that optimizes the point clouds and maintain the characteristics of the initial data. This method detects the features regions that describe the geometry of the surface. These features regions are detected using the saliency degree of vertices. Then, we define features points in each feature region and remove redundant vertices. Finally, we will show the robustness of our methodviadifferent experimental results. Moreover, we will study the stability of our method according to noise.

scholarly journals BRUTE FORCE MATCHING BETWEEN CAMERA SHOTS AND SYNTHETIC IMAGES FROM POINT CLOUDS

2016 ◽  
Vol XLI-B5 ◽  
pp. 771-777 ◽  
Author(s):  
R. Boerner ◽  
M. Kröhnert
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Spatial Information ◽  
Point Clouds ◽  
Synthetic Image ◽  
3D Point Cloud ◽  
Point Cloud Data ◽  
Cloud Data ◽  
Force Matching ◽  
3D Point Clouds

3D point clouds, acquired by state-of-the-art terrestrial laser scanning techniques (TLS), provide spatial information about accuracies up to several millimetres. Unfortunately, common TLS data has no spectral information about the covered scene. However, the matching of TLS data with images is important for monoplotting purposes and point cloud colouration. Well-established methods solve this issue by matching of close range images and point cloud data by fitting optical camera systems on top of laser scanners or rather using ground control points. &lt;br&gt;&lt;br&gt; The approach addressed in this paper aims for the matching of 2D image and 3D point cloud data from a freely moving camera within an environment covered by a large 3D point cloud, e.g. a 3D city model. The key advantage of the free movement affects augmented reality applications or real time measurements. Therefore, a so-called real image, captured by a smartphone camera, has to be matched with a so-called synthetic image which consists of reverse projected 3D point cloud data to a synthetic projection centre whose exterior orientation parameters match the parameters of the image, assuming an ideal distortion free camera.

The Research of Aeroplane Engine Blade 3D Point Clouds Processing

2012 ◽  
Vol 241-244 ◽  
pp. 2129-2132
Author(s):  
Hao Wang ◽  
Dong Yan Wang ◽  
Ting Jian Dong ◽  
Tao Wang
Keyword(s):  
Point Cloud ◽  
Measuring Equipment ◽  
Point Clouds ◽  
Point Cloud Data ◽  
Machining Time ◽  
Cloud Data ◽  
3D Point Clouds ◽  
Engine Blade ◽  
Reduced Modeling ◽  
Cloud Data Processing

This paper made the point cloud data processing for the aircraft engine’s blade. First, collected rough point cloud data by using visual measuring equipment. Then, noise reduced and smoothed, feature detected the point cloud data, took the reasonable simplification, finished pre-processing the point cloud data. Finally, took the surface fitting for the point cloud data after processed. The result proved that processing the point cloud data reduced modeling and machining time, and improved smoothness of the model.

scholarly journals INTEGRATION OF 3D POINT CLOUDS WITH SEMANTIC 3D CITY MODELS – PROVIDING SEMANTIC INFORMATION BEYOND CLASSIFICATION

2021 ◽  
Vol VIII-4/W2-2021 ◽  
pp. 105-112
Author(s):  
C. Beil ◽  
T. Kutzner ◽  
B. Schwab ◽  
B. Willenborg ◽  
A. Gawronski ◽  
...  
Keyword(s):  
Point Cloud ◽  
Semantic Information ◽  
Spatial Information ◽  
Point Clouds ◽  
Semantic Knowledge ◽  
Point Cloud Data ◽  
Cloud Data ◽  
3D City Models ◽  
3D Point Clouds ◽  
City Models

Abstract. A range of different and increasingly accessible acquisition methods, the possibility for frequent data updates of large areas, and a simple data structure are some of the reasons for the popularity of three-dimensional (3D) point cloud data. While there are multiple techniques for segmenting and classifying point clouds, capabilities of common data formats such as LAS for providing semantic information are mostly limited to assigning points to a certain category (classification). However, several fields of application, such as digital urban twins used for simulations and analyses, require more detailed semantic knowledge. This can be provided by semantic 3D city models containing hierarchically structured semantic and spatial information. Although semantic models are often reconstructed from point clouds, they are usually geometrically less accurate due to generalization processes. First, point cloud data structures / formats are discussed with respect to their semantic capabilities. Then, a new approach for integrating point clouds with semantic 3D city models is presented, consequently combining respective advantages of both data types. In addition to elaborate (and established) semantic concepts for several thematic areas, the new version 3.0 of the international Open Geospatial Consortium (OGC) standard CityGML also provides a PointCloud module. In this paper a scheme is shown, how CityGML 3.0 can be used to provide semantic structures for point clouds (directly or stored in a separate LAS file). Methods and metrics to automatically assign points to corresponding Level of Detail (LoD)2 or LoD3 models are presented. Subsequently, dataset examples implementing these concepts are provided for download.

scholarly journals Performance analysis of different classifiers in segmenting point cloud data

2021 ◽  
Vol 2107 (1) ◽  
pp. 012003
Author(s):  
N I Boslim ◽  
S A Abdul Shukor ◽  
S N Mohd Isa ◽  
R Wong
Keyword(s):  
Random Forest ◽  
Point Cloud ◽  
Laser Scanner ◽  
Point Clouds ◽  
Nearest Neighbour ◽  
Point Cloud Data ◽  
Historical Building ◽  
Cloud Data ◽  
3D Point Clouds ◽  
Different Types

Abstract 3D point clouds are a set of point coordinates that can be obtained by using sensing device such as the Terrestrial Laser Scanner (TLS). Due to its high capability in collecting data and produce a strong density point cloud surrounding it, segmentation is needed to extract information from the massive point cloud containing different types of objects, apart from the object of interest. Bell Tower of Tawau, Sabah has been chosen as the object of interest to study the performance of different types of classifiers in segmenting the point cloud data. A state-of-the-art TLS was used to collect the data. This research’s aim is to segment the point cloud data of the historical building from its scene by using two different types of classifier and to study their performances. Two main classifiers commonly used in segmenting point cloud data of interest like building are tested here, which is Random Forest (RF) and k-Nearest Neighbour (kNN). As a result, it is found out that Random Forest classifier performs better in segmenting the existing point cloud data that represent the historic building compared to k-Nearest Neighbour classifier.

scholarly journals BRUTE FORCE MATCHING BETWEEN CAMERA SHOTS AND SYNTHETIC IMAGES FROM POINT CLOUDS

2016 ◽  
Vol XLI-B5 ◽  
pp. 771-777
Author(s):  
R. Boerner ◽  
M. Kröhnert
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Spatial Information ◽  
Point Clouds ◽  
Synthetic Image ◽  
3D Point Cloud ◽  
Point Cloud Data ◽  
Cloud Data ◽  
Force Matching ◽  
3D Point Clouds

3D point clouds, acquired by state-of-the-art terrestrial laser scanning techniques (TLS), provide spatial information about accuracies up to several millimetres. Unfortunately, common TLS data has no spectral information about the covered scene. However, the matching of TLS data with images is important for monoplotting purposes and point cloud colouration. Well-established methods solve this issue by matching of close range images and point cloud data by fitting optical camera systems on top of laser scanners or rather using ground control points. <br><br> The approach addressed in this paper aims for the matching of 2D image and 3D point cloud data from a freely moving camera within an environment covered by a large 3D point cloud, e.g. a 3D city model. The key advantage of the free movement affects augmented reality applications or real time measurements. Therefore, a so-called real image, captured by a smartphone camera, has to be matched with a so-called synthetic image which consists of reverse projected 3D point cloud data to a synthetic projection centre whose exterior orientation parameters match the parameters of the image, assuming an ideal distortion free camera.

scholarly journals Performance Evaluation of Robot Localization Using 2D and 3D Point Clouds

2017 ◽  
Vol 29 (5) ◽  
pp. 928-934
Author(s):  
Kiyoaki Takahashi ◽  
◽  
Takafumi Ono ◽  
Tomokazu Takahashi ◽  
Masato Suzuki ◽  
...  
Keyword(s):  
Mobile Robots ◽  
Point Cloud ◽  
Image Data ◽  
Point Clouds ◽  
Autonomous Mobile Robots ◽  
Point Cloud Data ◽  
Cloud Data ◽  
3D Point Clouds ◽  
Laser Range Finders ◽  
2D And 3D

Autonomous mobile robots need to acquire surrounding environmental information based on which they perform their self-localizations. Current autonomous mobile robots often use point cloud data acquired by laser range finders (LRFs) instead of image data. In the virtual robot autonomous traveling tests we have conducted in this study, we have evaluated the robot’s self-localization performance on Normal Distributions Transform (NDT) scan matching. This was achieved using 2D and 3D point cloud data to assess whether they perform better self-localizations in case of using 3D or 2D point cloud data.

Sign in / Sign up

Export Citation Format

Share Document