Submit to Remote Sensing Review for Remote Sensing Propose a Special Issue

Journal Browser

► Journal Browser

Advances in Scene Understanding, 3D Reconstruction and Simulation for Autonomous Things

Special Issue Editors
Special Issue Information
Keywords
Published Papers

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

Deadline for manuscript submissions: closed (31 December 2022) | Viewed by 9177

Share This Special Issue

Special Issue Editors

Prof. Dr. Kyungeun Cho

E-Mail Website
Guest Editor

Department of Multimedia Engineering, Dongguk University, Seoul, Korea
Interests: 3D reconstruction; artificial intelligence for game and robot; virtual reality; NUI/NUX; human–robot interaction
Special Issues, Collections and Topics in MDPI journals

Prof. Dr. Pradip Kumar Sharma

E-Mail Website
Guest Editor

Department of Computing Science, University of Aberdeen, Aberdeen, UK
Interests: edge computing; IoT security; blockchain; software-defined networking; social networking
Special Issues, Collections and Topics in MDPI journals

Prof. Dr. Wei Song

E-Mail Website
Guest Editor

College of Computer Science and Technology, North China University of Technology, Beijing 100144, China
Interests: environment perception; unmanned ground vehicle; 3D reconstruction; object recognition
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Autonomous inventions independently operate and perform necessary functions through the perception of their surroundings, and the autonomous systems embedded in them are required to facilitate a response to external conditions, determining decisions to guarantee the completion of tasks. Although existing autonomous inventions, such as vehicles, robots, and drones, can execute tasks safely in simple scenarios—such as self-driving, indoor path planning, and indoor/outdoor scene scanning and reconstructing for virtual reality (VR)/augmented reality (AR)/mixed reality (MR) applications, etc., their accuracy is limited in more complex situations. Several external variables, such as object occlusion, hazy weather, adverse illumination, visual clutter, unrecognized objects, insufficient training data, etc., are responsible for this lapse in accuracy. The widespread, less-situational use of autonomous inventions requires advances in a range of technologies, including data obtained through sensors such as cameras or LiDAR, needing to be analyzed and accurately interpreted. During this process, semantic scene segmentation and object detection, recognition, and tracking are the fundamental methods needed for environmental perception in autonomous inventions, with the possibility of independently utilizing or integrating data presented in RGB images, depth images, and point clouds, this data, captured in multiple frames or views, is able to be merged to achieve more reliable results. In the case of autonomous vehicles, the 3D reconstruction of surrounding environments includes information regarding the volume of objects, which facilitates the path-planning task. The development of autonomous inventions also faces the the lack of ground-truth data. To handle this problem, simulation systems have been developed to simulate various environmental variables, including extraneous objects, weather conditions, and illumination, thereby providing accurate ground-truth data for deep learning-based methods.

This collection is a continuation and extension of the first Special Issue, “3D Reconstruction and Visualization of Dynamic Object/Scenes Using Data Fusion”, this time including more topics related to autonomous inventions, especially those related to sensor data processing, analysis, scene understanding, 3D reconstruction, multi-sensor data fusion, and simulation techniques and systems.

Prof. Dr. Kyungeun Cho
Prof. Dr. Pradip Kumar Sharma
Prof. Dr. Wei Song
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 submissions that pass pre-check are 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 2700 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

image understanding
image interpretation
image recognition
image analysis
pattern analysis
multimodal data fusion
object detection and tracking
scene understanding
3D scene reconstruction
point cloud denoising
3D object recognition
AR/VR/MR/XR
scenario extraction from videos
understanding and representation of simulation
understanding and representation of scenarios
multiscale representation of simulation
simulation acceleration approach
inverse model-based simulation
physics-based models from data
sensor-realistic simulation in real time
photorealistic simulation

Published Papers (4 papers)

Download All Papers

Research

19 pages, 14472 KiB

Open AccessArticle

A Rapid Water Region Reconstruction Scheme in 3D Watershed Scene Generated by UAV Oblique Photography

by Yinguo Qiu, Yaqin Jiao, Juhua Luo, Zhenyu Tan, Linsheng Huang, Jinling Zhao, Qitao Xiao and Hongtao Duan

Remote Sens. 2023, 15(5), 1211; https://doi.org/10.3390/rs15051211 - 22 Feb 2023

Cited by 5 | Viewed by 1815

Abstract

Oblique photography technology based on UAV (unmanned aerial vehicle) provides an effective means for the rapid, real-scene 3D reconstruction of geographical objects on a watershed scale. However, existing research cannot achieve the automatic and high-precision reconstruction of water regions due to the sensitivity of water surface patterns to wind and waves, reflections of objects on the shore, etc. To solve this problem, a novel rapid reconstruction scheme for water regions in 3D models of oblique photography is proposed in this paper. It extracts the boundaries of water regions firstly using a designed eight-neighborhood traversal algorithm, and then reconstructs the triangulated irregular network (TIN) of water regions. Afterwards, the corresponding texture images of water regions are intelligently selected and processed using a designed method based on coordinate matching, image stitching and clipping. Finally, the processed texture images are mapped to the obtained TIN, and the real information about water regions can be reconstructed, visualized and integrated into the original real-scene 3D environment. Experimental results have shown that the proposed scheme can rapidly and accurately reconstruct water regions in 3D models of oblique photography. The outcome of this work can refine the current technical system of 3D modeling by UAV oblique photography and expand its application in the construction of twin watershed, twin city, etc. Full article

(This article belongs to the Special Issue Advances in Scene Understanding, 3D Reconstruction and Simulation for Autonomous Things)

► Show Figures

Graphical abstract

30 pages, 150083 KiB

Open AccessArticle

Deep-Learning-Based Three-Dimensional Detection of Individual Wave-Dissipating Blocks from As-Built Point Clouds Measured by UAV Photogrammetry and Multibeam Echo-Sounder

by Yajun Xu, Satoshi Kanai, Hiroaki Date and Tomoaki Sano

Remote Sens. 2022, 14(21), 5575; https://doi.org/10.3390/rs14215575 - 4 Nov 2022

Cited by 3 | Viewed by 1989

Abstract

Wave-dissipating blocks are the armor elements of breakwaters that protect beaches, ports, and harbors from erosion by waves. Monitoring the poses of individual wave-dissipating blocks benefits the accuracy of the block supplemental work plan, recording of the construction status, and monitoring of long-term pose change in blocks. This study proposes a deep-learning-based approach to detect individual blocks from large-scale three-dimensional point clouds measured with a pile of wave-dissipating blocks placed overseas and underseas using UAV photogrammetry and a multibeam echo-sounder. The approach comprises three main steps. First, the instance segmentation using our originally designed deep convolutional neural network partitions an original point cloud into small subsets of points, each corresponding to an individual block. Then, the block-wise 6D pose is estimated using a three-dimensional feature descriptor, point cloud registration, and CAD models of blocks. Finally, the type of each segmented block is identified using model registration results. The results of the instance segmentation on real-world and synthetic point cloud data achieved 70–90% precision and 50–76% recall with an intersection of union threshold of 0.5. The pose estimation results on synthetic data achieved 83–95% precision and 77–95% recall under strict pose criteria. The average block-wise displacement error was 30 mm, and the rotation error was less than

2^{\circ}

. The pose estimation results on real-world data showed that the fitting error between the reconstructed scene and the scene point cloud ranged between 30 and 50 mm, which is below 2% of the detected block size. The accuracy in the block-type classification on real-world point clouds reached about 95%. These block detection performances demonstrate the effectiveness of our approach. Full article

(This article belongs to the Special Issue Advances in Scene Understanding, 3D Reconstruction and Simulation for Autonomous Things)

► Show Figures

Graphical abstract

18 pages, 2631 KiB

Open AccessArticle

DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV

by Wei Song, Zhen Liu, Ying Guo, Su Sun, Guidong Zu and Maozhen Li

Remote Sens. 2022, 14(15), 3825; https://doi.org/10.3390/rs14153825 - 8 Aug 2022

Cited by 3 | Viewed by 2029

Abstract

Semantic segmentation in LiDAR point clouds has become an important research topic for autonomous driving systems. This paper proposes a dynamic graph convolution neural network for LiDAR point cloud semantic segmentation using a polar bird’s-eye view, referred to as DGPolarNet. LiDAR point clouds are converted to polar coordinates, which are rasterized into regular grids. The points mapped onto each grid distribute evenly to solve the problem of the sparse distribution and uneven density of LiDAR point clouds. In DGPolarNet, a dynamic feature extraction module is designed to generate edge features of perceptual points of interest sampled by the farthest point sampling and K-nearest neighbor methods. By embedding edge features with the original point cloud, local features are obtained and input into PointNet to quantize the points and predict semantic segmentation results. The system was tested on the Semantic KITTI dataset, and the segmentation accuracy reached 56.5% Full article

(This article belongs to the Special Issue Advances in Scene Understanding, 3D Reconstruction and Simulation for Autonomous Things)

► Show Figures

Figure 1

17 pages, 4604 KiB

Open AccessArticle

2D&3DHNet for 3D Object Classification in LiDAR Point Cloud

by Wei Song, Dechao Li, Su Sun, Lingfeng Zhang, Yu Xin, Yunsick Sung and Ryong Choi

Remote Sens. 2022, 14(13), 3146; https://doi.org/10.3390/rs14133146 - 30 Jun 2022

Cited by 12 | Viewed by 2269

Abstract

Accurate semantic analysis of LiDAR point clouds enables the interaction between intelligent vehicles and the real environment. This paper proposes a hybrid 2D and 3D Hough Net by combining 3D global Hough features and 2D local Hough features with a classification deep learning network. Firstly, the 3D object point clouds are mapped into the 3D Hough space to extract the global Hough features. The generated global Hough features are input into the 3D convolutional neural network for training global features. Furthermore, a multi-scale critical point sampling method is designed to extract critical points in the 2D views projected from the point clouds to reduce the computation of redundant points. To extract local features, a grid-based dynamic nearest neighbors algorithm is designed by searching the neighbors of the critical points. Finally, the two networks are connected to the full connection layer, which is input into fully connected layers for object classification. Full article

(This article belongs to the Special Issue Advances in Scene Understanding, 3D Reconstruction and Simulation for Autonomous Things)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Advances in Scene Understanding, 3D Reconstruction and Simulation for Autonomous Things

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (4 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI