Special Issue "Point Cloud Processing in Remote Sensing Technology"

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Remote Sensing Image Processing".

Deadline for manuscript submissions: 30 November 2022.

Special Issue Editors

Dr. Johannes Otepka
E-Mail Website
Guest Editor
TU Wien, Department of Geodesy and Geoinformation, Wiedner Hauptstraße 8/E120, 1040 Vienna, Austria
Interests: point cloud processing; laser scanning; spatial indices; efficient processing concepts; least-squares; point cloud orientation and strip adjustment
Dr. Martin Weinmann
E-Mail Website
Guest Editor
Institute of Photogrammetry and Remote Sensing, Karlsruhe Institute of Technology (KIT), Englerstr. 7, 76131 Karlsruhe, Germany
Interests: computer vision; pattern recognition; machine learning; photogrammetry; remote sensing
Special Issues and Collections in MDPI journals
Dr. Di Wang
E-Mail Website
Guest Editor
Department of Remote Sensing Science and Technology, School of Electronic Engineering, Xidian University, South Taibai Road 2, Xi'an 710071, China
Interests: LiDAR remote sensing; point cloud processing; 3D reconstruction; tree modeling; vegetation structure analysis
Dr. Kourosh Khoshelham
E-Mail Website
Guest Editor
Department of Infrastructure Engineering, The University of Melbourne, Melbourne, VIC 3010, Australia
Interests: photogrammetry; 3D computer vision; remote sensing; machine learning; deep learning; automated interpretation of imagery and point clouds
Special Issues and Collections in MDPI journals

Special Issue Information

Dear Colleagues,

Modern data acquisition with active or passive remote sensing techniques often results in 3D point clouds. While point clouds were long regarded as an intermediate product for deriving 2.5D or 3D models, they are nowadays accepted as a primary data product that plays a central role in a huge variety of applications.

The technological advances and the miniaturisation of remote sensing hardware led to the development of a large number of distinctive devices for capturing 3D point clouds at different scales, resolutions and precisions. For instance, laser scanners, single-camera systems and multi-camera systems (in conjunction with image matching), RGBD cameras, time-of-flight sensors, synthetic aperture radar systems, ground-penetrating radar systems, echo sounding systems, index arms with tactile tip or scanning heads, etc., are used on static (e.g., tripod) or kinematic platforms (e.g., robot, car, boat, UAV, helicopter, airplane, or satellite) to capture objects or scenes of different scale via close-range, mid-range or far-range measurements.

Although the capturing procedure is the starting point for many applications, the processing of 3D point clouds is essential to visualise, enrich, analyse, quantify, evaluate, model, and to understand the measured object or scene. A processing pipeline typically consists of multiple stages, such as point cloud orientation, co-registration, quality control, feature extraction, semantic segmentation and classification, object detection and recognition, change detection, and object modelling. This Special Issue will report cutting-edge methods, algorithms, and data structures of certain stages or comprehensive processing pipelines for specific applications or sensors.

The Special Issue invites authors to submit contributions in (but not limited to) the following topics:

  • Point cloud generation and quality analyses for new or improved sensors, such as miniaturised cameras and laser scanners, integrated sensors, Geigermode and single-photon LiDAR systems, UAV-based laser scanning systems, mobile mapping systems, multi-beam echo sounding systems, tomographic synthetic aperture radar systems, etc.;
  • Deep learning methods, specific network designs, transfer learning, and data organisation strategies for realising new or improved classification and object detection tasks as required for self-driving cars, indoor navigation, object modelling, etc.;
  • Classical semantic segmentation and classification methods are still relevant for many tasks, especially when processing huge point clouds due to the computational burden of deep learning and the lack of a sufficient amount of training data;
  • Innovative 2.5D and 3D modelling algorithms, as often used in mobile and corridor mapping, but also for traditional topographic point clouds, such as terrain, surface, building and tree modelling;
  • Data fusion of point clouds acquired from different sensors, scales, and accuracies. Today's sensor heads typically combine multiple sensor elements, such as differently oriented cameras (forward, nadir, backward and oblique views), laser scanners with multiple channels or different wavelengths (infrared and green laser diodes), etc.;
  • Multi-temporal analyses, which are used, for example, for change detection, updating inventory databases, land slide monitoring, or disaster management;
  • Methods and algorithms for interacting with point clouds to visualise, inspect, and highlight specific aspects of the dataset;
  • Optimised algorithms, strategies and data structures for efficiently processing huge point clouds.

Dr. Johannes Otepka
Dr. Martin Weinmann
Dr. Di Wang
Prof. Kourosh Khoshelham
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Point cloud generation
  • Mobile mapping
  • LiDAR
  • Photogrammetric point clouds
  • Dense image matching
  • 3D modelling
  • Point cloud analysis
  • Quality and accuracy estimation
  • Feature extraction
  • Semantic segmentation
  • Supervised and unsupervised machine learning
  • Deep learning
  • Data fusion
  • Change detection

Published Papers (2 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Article
A LiDAR/Visual SLAM Backend with Loop Closure Detection and Graph Optimization
Remote Sens. 2021, 13(14), 2720; https://0-doi-org.brum.beds.ac.uk/10.3390/rs13142720 - 10 Jul 2021
Cited by 2 | Viewed by 620
Abstract
LiDAR (light detection and ranging), as an active sensor, is investigated in the simultaneous localization and mapping (SLAM) system. Typically, a LiDAR SLAM system consists of front-end odometry and back-end optimization modules. Loop closure detection and pose graph optimization are the key factors [...] Read more.
LiDAR (light detection and ranging), as an active sensor, is investigated in the simultaneous localization and mapping (SLAM) system. Typically, a LiDAR SLAM system consists of front-end odometry and back-end optimization modules. Loop closure detection and pose graph optimization are the key factors determining the performance of the LiDAR SLAM system. However, the LiDAR works at a single wavelength (905 nm), and few textures or visual features are extracted, which restricts the performance of point clouds matching based loop closure detection and graph optimization. With the aim of improving LiDAR SLAM performance, in this paper, we proposed a LiDAR and visual SLAM backend, which utilizes LiDAR geometry features and visual features to accomplish loop closure detection. Firstly, the bag of word (BoW) model, describing the visual similarities, was constructed to assist in the loop closure detection and, secondly, point clouds re-matching was conducted to verify the loop closure detection and accomplish graph optimization. Experiments with different datasets were carried out for assessing the proposed method, and the results demonstrated that the inclusion of the visual features effectively helped with the loop closure detection and improved LiDAR SLAM performance. In addition, the source code, which is open source, is available for download once you contact the corresponding author. Full article
(This article belongs to the Special Issue Point Cloud Processing in Remote Sensing Technology)
Show Figures

Graphical abstract

Article
Hyperspectral LiDAR-Based Plant Spectral Profiles Acquisition: Performance Assessment and Results Analysis
Remote Sens. 2021, 13(13), 2521; https://0-doi-org.brum.beds.ac.uk/10.3390/rs13132521 - 28 Jun 2021
Viewed by 553
Abstract
In precision agriculture, efficient fertilization is one of the most important pursued goals. Vegetation spectral profiles and the corresponding spectral parameters are usually employed for vegetation growth status indication, i.e., vegetation classification, bio-chemical content mapping, and efficient fertilization guiding. In view of the [...] Read more.
In precision agriculture, efficient fertilization is one of the most important pursued goals. Vegetation spectral profiles and the corresponding spectral parameters are usually employed for vegetation growth status indication, i.e., vegetation classification, bio-chemical content mapping, and efficient fertilization guiding. In view of the fact that the spectrometer works by relying on ambient lighting condition, hyperspectral/multi-spectral LiDAR (HSL/MSL) was invented to collect the spectral profiles actively. However, most of the HSL/MSL works with the wavelength specially selected for specific applications. For precision agriculture applications, a more feasible HSL capable of collecting spectral profiles at wide-range spectral wavelength is necessary to extract various spectral parameters. Inspired by this, in this paper, we developed a hyperspectral LiDAR (HSL) with 10 nm spectral resolution covering 500~1000 nm. Different vegetation leaf samples were scanned by the HSL, and it was comprehensively assessed for wide-range wavelength spectral profiles acquirement, spectral parameters extraction, vegetation classification, and the laser incident angle effect. Specifically, three experiments were carried out: (1) spectral profiles results were compared with that from a SVC spectrometer (HR-1024, Spectra Vista Corporation); (2) the extracted spectral parameters from the HSL were assessed, and they were employed as the input features of a support vector machine (SVM) classifier with multiple labels to classify the vegetation; (3) in view of the influence of the laser incident angle on the HSL reflected laser intensities, we analyzed the laser incident angle effect on the spectral parameters values. The experimental results demonstrated the developed HSL was more feasible for acquiring spectral profiles with wide-range wavelength, and spectral parameters and vegetation classification results also indicated its great potentials in precision agriculture application. Full article
(This article belongs to the Special Issue Point Cloud Processing in Remote Sensing Technology)
Show Figures

Graphical abstract

Back to TopTop