3D Instance Segmentation and Object Detection Framework Based on the Fusion of Lidar Remote Sensing and Optical Image Sensing

Bai, Ling; Li, Yinguo; Cen, Ming; Hu, Fangchao

doi:10.3390/rs13163288

Cited by 15 publications

(6 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…# The object ID output by the point cloud sensor module integrated into MATLAB/Simulink is used to determine whether the collected point is reflected from the TV body. Note that many state-of-the-art object detection and semantic segmentation approaches could reliably distinguish the target vehicle from surrounding objects, pavements, and roadway environment in the real world [68,69]. Thus, although without the investigation of this actual procedure in this study, its completion could be simulated via the aforementioned method.…”

Section: Methodsmentioning

confidence: 99%

Examining the feasibility of current spiral curve design controls for LiDAR‐based automated vehicles

Wang

Mao

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Enabling large-scale deployment of automated vehicles (AVs) in the near future requires answering the question firstly: whether AVs could safely adapt to as-built road geometry? This study aims to examine the feasibility of current spiral curve design controls for LiDAR-based AVs (LAVs) from the perspective of the available sight distance (ASD). A series of tests featuring the design speed (V d ), lengths of tangent (L T ) and spiral (L S ), circular curve radii (R), and point thresholds for detection (N T ) were simulated in PreScan/MATLAB/Simulink co-simulation platform. The ASDs affected by those parameters' combined effects were analyzed and compared with required stopping sight distances (RSDs) of human-driven vehicles (HVs) and level 3 to 5 (L3-L5) LAVs followed by proposing the ASD-oriented safe speeds and the corresponding speed limits. The results indicate that: (1) the combination of the tangent-spiral curve-circular curve causes a shorter ASD than that without the spiral curve; (2) a longer spiral curve causes a shorter ASD; and (3) only a low-type combination of R, L S , V d conditions is feasible for L3 LAVs while L4 or L5 LAVs have difficulties in dealing with high-type conditions. These findings help understand the ASD for AVs and provide safety-critical speed references for administrators. INTRODUCTIONAutomated vehicles (AVs), generally referring to vehicles equipped with level 3 to 5 automation systems [1], are anticipated to serve as an optimal solution for improving traffic safety mainly by mitigating driver errors [2-4]. For example, AVs are capable of obtaining more stable perception-functional performances and response ability through their sensor-based perception systems and onboard computing units [5], while the perception system of human drivers relying on the eyes and brains is easily affected by their psychophysical conditionsThis is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

show abstract

Section: Methodsmentioning

confidence: 99%

Examining the feasibility of current spiral curve design controls for LiDAR‐based automated vehicles

Wang

Mao

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

show abstract

“…Feng [23] proposes MigrationNet, a learning-based method for locating and visualizing subsurface objects. Ling et al [24] propose a 3D object detection framework for urban transportation scenes based on the fusion of Lidar remote sensing technology and optical image sensing technology. To extract the hyperbolic features from GPR B-scan images, Faster R-CNN [25,26], a classic 2D object detection algorithm, is employed.…”

Section: Related Workmentioning

confidence: 99%

MV-GPRNet: Multi-View Subsurface Defect Detection Network for Airport Runway Inspection Based on GPR

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

The detection and restoration of subsurface defects are essential for ensuring the structural reliability of airport runways. Subsurface inspections can be performed with the aid of a robot equipped with a Ground Penetrating Radar (GPR). However, interpreting GPR data is extremely difficult, as GPR data usually contains severe clutter interference. In addition, many different types of subsurface defects present similar features in B-scan images, making them difficult to distinguish. Consequently, this makes later maintenance work harder as different subsurface defects require different restoration measures. Thus, to automate the inspection process and improve defect identification accuracy, a novel deep learning algorithm, MV-GPRNet, is proposed. Instead of traditionally using GPR B-scan images only, MV-GPRNet utilizes multi-view GPR data to robustly detect regions with defects despite significant interference. It originally fuses the 3D feature map in C-scan data and the 2D feature map in Top-scan data for defect classification and localization. With our runway inspection robot, a large number of real runway data sets from three international airports have been used to extensively test our method. Experimental results indicate that the proposed MV-GPRNet outperforms state-of-the-art (SOTA) approaches. In particular, MV-GPRNet achieves F1 measurements for voids, cracks, subsidences, and pipes at 91%, 69%, 90%, and 100%, respectively.

show abstract

“…The rapid advancement of science and technology has made remote-sensing image technology indispensable for various applications. Some examples of these applications include monitoring for diseases, transportation planning, environmental monitoring, crop harvest analysis, geological surveys, and identifying objects used in military operations [1][2][3][4][5][6].…”

Section: Introductionmentioning

confidence: 99%

Small Object Detection Based on Deep Learning for Remote Sensing: A Comprehensive Review

Wang

Song

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

With the accelerated development of artificial intelligence, remote-sensing image technologies have gained widespread attention in smart cities. In recent years, remote sensing object detection research has focused on detecting and counting small dense objects in large remote sensing scenes. Small object detection, as a branch of object detection, remains a significant challenge in research due to the image resolution, size, number, and orientation of objects, among other factors. This paper examines object detection based on deep learning and its applications for small object detection in remote sensing. This paper aims to provide readers with a thorough comprehension of the research objectives. Specifically, we aggregate the principal datasets and evaluation methods extensively employed in recent remote sensing object detection techniques. We also discuss the irregularity problem of remote sensing image object detection and overview the small object detection methods in remote sensing images. In addition, we select small target detection methods with excellent performance in recent years for experiments and analysis. Finally, the challenges and future work related to small object detection in remote sensing are highlighted.

show abstract

3D Instance Segmentation and Object Detection Framework Based on the Fusion of Lidar Remote Sensing and Optical Image Sensing

Cited by 15 publications

References 42 publications

Examining the feasibility of current spiral curve design controls for LiDAR‐based automated vehicles

Examining the feasibility of current spiral curve design controls for LiDAR‐based automated vehicles

MV-GPRNet: Multi-View Subsurface Defect Detection Network for Airport Runway Inspection Based on GPR

Small Object Detection Based on Deep Learning for Remote Sensing: A Comprehensive Review

Contact Info

Product

Resources

About