Light Detection And Ranging Scanning Research Articles

Phase-Modulated Continuous-Wave Coherent Ranging Method and Anti-Interference Evaluation

Light detection and ranging (LiDAR) has been applied in many areas because of its excellent performance. An easily achievable, cost-effective, and high-performance ranging method is a major challenge of LiDAR. Meanwhile, with the increasing applications of LiDAR, numerous LiDARs can be made to operate simultaneously, and potential interference is inevitable. Therefore, immunity against interference is paramount in LiDAR systems. In this paper, we demonstrated a ranging method referred to as phase-modulated continuous-wave (PhMCW). A detection range of 50 m and a ranging error of 2.2 cm are achieved. A one-dimensional scanning LiDAR system that is capable of detecting targets at 28 m is built, demonstrating the validation of the PhMCW method. Moreover, we propose a quantitative method for evaluating the anti-interference capability of lidar systems. The p-values of the Ljung–Box test were 0.0589 and 0.6327 for ToF and coherent LiDAR interferences, respectively, indicating that the PhMCW system is immune to interference. The proposed method can be applied to all types of LiDAR systems, regardless of the ranging method or beam-steering technique used.

AFLI-Calib: Robust LiDAR-IMU extrinsic self-calibration based on adaptive frame length LiDAR odometry

As an effective complement to common laser scanning systems, the portable laser scanning system can acquire point clouds flexibly and quickly. Calibration between Light detection and ranging (LiDAR) sensors and inertial measurement units (IMU) is the prerequisite for laser scanning systems to obtain high-quality point clouds. Related methods have been proposed in the last two decades, where the global navigation satellite system (GNSS) or high-precision calibration fields are commonly used. However, the extrinsic self-calibration of LiDAR-IMU is challenging, due to the large distortion in single-frame point cloud caused by rapid motion and the position errors of IMU integration which drift quickly. At the same time, the highly dynamic motion patterns of portable devices and the changes in the scanned scene structure are not well considered in existing LiDAR odometry methods. To take better advantage of the characteristics of non-repetitive scanning LiDAR sensor, this paper proposes AFLI-Calib, which utilizes adaptive frame length LiDAR odometry to perform the extrinsic self-calibration of LiDAR-IMU. Unlike LiDAR odometry methods with a fixed frame length, the LiDAR frame length is dynamically adjusted according to the motion state of sensors and the matching stability of scenes. The single-frame point cloud is registered to the map through a linear-based continuous-time model, eliminating the motion distortion correction in advance. For further optimization of trajectory and extrinsic parameters, IMU raw measurements and LiDAR observations are involved in the multi-constraint optimization, through tightly-coupled IMU pre-integration constraints, LiDAR point-to-plane constraints, and prior constraints. The method is fully validated using self-collected calibration data of indoor and outdoor scenes and different motion modes. Experiments show that on the test data, the translation parameter accuracy of the method is 0.041 m, which is 56.3% higher than the state-of-the-art method. The standard deviation is significantly reduced, with translation deviation (0.017 m, 0.024 m, 0.022 m) and rotation deviation (0.17°, 0.25°, 0.15°), which verifies the robustness of our method. The average RMSE of distances to the reference point cloud acquired by the terrestrial laser scanning system (TLS) is 0.042 m, showing a high accuracy calibration result. Comparative experiments with the fixed frame length LiDAR odometry method and classical “correction-then-registration” motion distortion model further verify the superiority and effectiveness of the proposed adaptive frame length LiDAR odometry.

Feasibility of Low-Cost LiDAR Scanner Implementation in Forest Sampling Techniques

Despite the growing impact of remote sensing technology in forest inventories globally, there is a continuous need for ground measurements on sample plots. Even though the newest volume assessment methodology requires fewer sample plots, the accuracy of ground-recorded data influences the final accuracy of forest stand modeling. Therefore, effective and economically justified tools are in the continuous interest of foresters. In the presented research, a consumer-grade light detection and ranging (LiDAR) sensor mounted on iPad was used for forest inventory sample plot data collection—including tree location and diameter breast height. In contrast to other similar research, feasibility and user-friendliness were also documented and emphasized. The study was conducted in 63 real sample plots used for the inventory of Polish forests. In total, 776 trees were scanned in 3 types of forest stands: pine, birch, and oak. The root mean square error was 0.28 m for tree locations and 0.06 m for diameter breast height. Various additional analyses were performed to describe the usage of an iPad in tree inventories. It was contended that low-cost LiDAR scanners might be successfully used in real forest conditions and can be considered a reliable and easy-to-implement tool in forest inventory measurements.

The effects of sampling and instrument orientation on LiDAR data from crop plots.

Wheat is one of the most widely consumed grains in the world and improving its yield, especially under severe climate conditions, is of great importance to world food security. Phenotyping methods can evaluate plants according to their different traits, such as yield and growth characteristics. Assessing the vertical stand structure of plants can provide valuable information about plant productivity and processes, mainly if this trait can be tracked throughout the plant's growth. Light Detection And Ranging (LiDAR) is a method capable of gathering three-dimensional data from wheat field trials and is potentially suitable for providing non-destructive, high-throughput estimations of the vertical stand structure of plants. The current study considers LiDAR and focuses on investigating the effects of sub-sampling plot data and data collection parameters on the canopy vertical profile (CVP). The CVP is a normalized, ground-referenced histogram of LiDAR point cloud data representing a plot or other spatial domain. The effects of sub-sampling of plot data, the angular field of view (FOV) of the LiDAR and LiDAR scan line orientation on the CVP were investigated. Analysis of spatial sub-sampling effects on CVP showed that at least 144000 random points (600 scan lines) or an area equivalent to three plants along the row were adequate to characterize the overall CVP of the aggregate plot. A comparison of CVPs obtained from LiDAR data for different FOV showed that CVPs varied with the angular range of the LiDAR data, with narrow ranges having a larger proportion of returns in the upper canopy and a lower proportion of returns in the lower part of the canopy. These findings will be necessary to establish minimum plot and sample sizes and compare data from studies where scan direction or field of view differ. These advancements will aid in making comparisons and inform best practices for using close-range LiDAR in phenotypic studies in crop breeding and physiology research.

Design of a circum-scanning light detection and ranging emission system

The need for imaging and detecting in various instrument areas has brought light detection and ranging (LiDAR) to the forefront of consumer technology. Among different LiDAR, microelectromechanical system (MEMS) LiDAR has more appeal due to its small size and high integration. However, due to its low resolution and slight scanning angle, MEMS LiDAR does not apply to all scenes. Thus we presented a 360-deg scanning LiDAR emission optical system. Design formulas were deduced from theoretical formulas. In this system, the aspheric lens collimated the beam, MEMS micromirrors tracked the concentric circular beams, and the converging lens compensated for the divergence of the outgoing beam. By doing so, the target was scanned 360 deg at high resolution. With Zemax, the fast-axis divergence angle was 0.2484 mrad, the slow-axis divergence angle was 0.1546 mrad, and the system energy utilization rate was 84.16% with an angular resolution of 0.0142 deg at a distance of 10,000 mm. We have provided a potential solution for improving the scanning angle and resolution of MEMS LiDAR.

Suitable LiDAR Platform for Measuring the 3D Structure of Mangrove Forests

Investigating the three-dimensional structure of mangrove forests is critical for their conservation and restoration. However, mangrove forests are difficult to survey in the field, and their 3D structure is poorly understood. Light detection and ranging (LiDAR) is considered an accurate and dependable method of measuring the 3D structure of mangrove forests. This study aimed to find a suitable LiDAR platform for obtaining attributes such as breast height diameter and canopy area, as well as for measuring a digital terrain model (DTM), the base data for hydrological analysis. A mangrove forest near the mouth of the Oura River in Aza-Oura, Nago City, Okinawa Prefecture, Japan, was studied. We used data from terrestrial LiDAR scanning “TLS” and unmanned aerial vehicle (UAV) LiDAR scanning “ULS” as well as data merged from TLS and ULS “Merge”. By interpolating point clouds of the ground surface, DTMs of 5 cm × 5 cm were created. DTMs obtained from ULS could not reproduce the heaps of Thalassina anomala or forest floor microtopography compared with those obtained from TLS. Considering that ULS had a few point clouds in the forest, automatic trunk identification could not be used to segment trees. TLS could segment trees by automatically identifying trunks, but the number of trees identified roughly doubled that of the visual identification results. The number of tree crowns identified using TLS and ULS was approximately one quarter of those identified visually, and many of them were larger in area than the visually traced crowns. The accuracy of tree segmentation using the canopy height model (CHM) was low. The number of canopy trees identified using Merge produced the best results, accounting for 61% of the visual identification results. Results of tree segmentation by CHM suggest that combining TLS and ULS measurements may improve tree canopy identification. Although ULS is a promising new technology, its applications are clearly limited, at least in mangrove forests such as the Oura River, where Bruguiera gymnorhiza is dominant. Depending on the application, using different LiDAR platforms, such as airborne LiDAR scanning, UAV LiDAR scanning, and TLS, is important. Merging 3D point clouds acquired by different platforms, as proposed in this study, is an important option in this case.

Characterization of Operational Vibrations of Steel-Girder Highway Bridges via LiDAR

This research is motivated by the need for rapidly deployable technologies such as wireless, non-contact or remote sensing for evaluating bridges under operating conditions to minimize the data collection time, avoid the disruption of traffic and increase the inspector’s safety. The objective established for this research is to explore the use of remote sensing (e.g., Light Detection and Ranging (LiDAR)) for characterizing the structural vibration of bridges to support and improve bridge assessment practices. To satisfy this objective, a field study was performed on a 12-span steel stringer bridge in the Philadelphia region. This structure was subjected to extensive LiDAR scanning and conventional vibration data collection through the use of accelerometers for validation purposes. The analysis of the data collected in the field revealed LiDAR’s capability for detecting the structure’s vibration. The field data displayed an error for LiDAR vs. accelerometers of between 1.9 and 10 percent. Additionally, numerical modeling was performed on MATLAB to allow for a better understanding of the interaction between the scanner and the structure. The numerical model presents a vibrating plate to represent a simply supported single-span bridge and a terrestrial LiDAR sensor located under the plate which scans while it is vibrating constantly without attenuation. Finally, a set of recommendations were established for the use of LiDAR scanning to evaluate the structure’s frequency of vibration.

Power Line Extraction and Reconstruction Methods from Laser Scanning Data: A Literature Review

Electricity has become an indispensable source of energy, and power lines play a crucial role in the functioning of modern societies. It is essential to inspect power lines promptly and precisely in order to ensure the safe and secure delivery of electricity. In steep and mountainous terrain, traditional surveying methods cannot inspect power lines precisely due to their nature. Remote sensing platforms, such as satellite and aerial images, thermal images, and light detection and ranging (LiDAR) points, were utilised for the detection and inspection of power lines. Nevertheless, with the advancements of remote sensing technologies, in recent years, LiDAR surveying has been favoured for power line corridor (PLC) inspection due to active and weather-independent nature of laser scanning. Laser ranging data and the precise location of the LiDAR can be used to generate a three-dimensional (3D) image of the PLC. The resulting 3D point cloud enables accurate extraction of power lines and measurement of their distances from the forest below. In the literature, there have been many proposals for power line extraction and reconstruction for PLC modelling. This article examines the pros and cons of each domain method, providing researchers involved in three-dimensional modelling of power lines using innovative LiDAR scanning systems with useful guidelines. To achieve these objectives, research papers were analysed, focusing primarily on geoscience-related journals and conferences for the extraction and reconstruction of power lines. There has been a growing interest in examining the extraction and reconstruction of power line spans with single and multi-conductor configurations using different image and point-based techniques. Our study provides a comprehensive overview of the methodologies offered by various approaches using laser scanning data from the perspective of power line extraction applications, as well as to discuss the benefits and drawbacks of each approach. The comparison revealed that, despite the tremendous potential of aerial and mobile laser scanning systems, human intervention and post-processing actions are still required to achieve the desired results. In addition, the majority of the methods have been evaluated on the small datasets, and very few methods have been focused on multi-conductor extraction and reconstruction for power lines modelling. These barriers hinder the automated extraction and reconstruction of power line using LiDAR data and point to unexplored areas for further research and serve as useful guidelines for future research directions. Several promising directions for future LiDAR experiments using deep learning methods are outlined in the hope that they will pave the way for applications of PLC modelling and assessment at a finer scale and on a larger scale.

Effects of laser scanner quality and tractor speed to characterise apple tree canopies

Precision orchard management requires non-destructive sensor technologies to identify key sources of variation in apple yield and quality so operations such as fruit thinning, root-pruning as well as harvesting can be better targeted and optimised. We compared the data quality of two light detection and ranging (LiDAR) scanners to assess their ability to estimate the volume of ‘Braeburn’ apple trees. Scanner one was a high-quality tractor-mounted LiDAR (SICK, LDMRS 400001, Düsseldorf, Germany) and the second system consisted of two lower-cost SICK LiDAR sensors fitted on a commercial orchard sprayer. To compare the two scanner systems, two tree row sections with different growth vigour were created by targeted root pruning to reduce annual shoot growth. LiDAR scans and reference assessments were made at the end of the growing season. In addition, LiDAR data were obtained at two tractor speeds: 1.5 and 7.0 km/h (standard spraying speed). Results show key canopy features like the beginning/end of rows or large gaps can be detected with the lower-cost sensors. However, tractor speed had a major effect on data quality. At 1.5 km/h, alpha-shaped tree volumes of the high-quality LiDAR show a high correlation (R2 = 0.84) with the reference assessments, whereas the data quality of the lower-cost sensors was not high enough to represent the tree volume with the alpha-shape algorithm (R2 = 0.09). Moreover, this work raises the question of how precise a sensor-based tree volume needs to be determined to describe heterogeneity in an orchard. Specifically, how to best match sensor data to manual methods of canopy measurement and allow sensor data to be used for multiple applications.

High-precision deformation analysis of yingxian wooden pagoda based on UAV image and terrestrial LiDAR point cloud

The monitoring of wooden pagodas is a very important task in the restoration of wooden pagodas. Traditionally, this labor has always been carried out by surveying personnel, who manually check all parts of the pagoda, which not only consumes huge manpower, but also suffers from low efficiency and measurement errors. This article evaluates the feasibility of combining portable 3D light detection and ranging (LiDAR) scanning and unmanned aerial vehicle (UAV) photogrammetry to perform these inspection tasks easily and accurately. The wooden pagoda's exterior picture and inside point cloud are acquired using a UAV and a LiDAR scanner, respectively. We propose a feature−based global alignment method to register the site point cloud. The error equation of the column of observed values is utilized as the beginning value of the feature constraint for global leveling. The beam method leveling model solves the spatial transformation parameters and the unknown point leveling values. Then, the Structure from Motion (SfM) algorithm of computer vision is used to realize the fusion of the dense point cloud of the exterior of the wooden pagoda generated from multiple non−measured images by global optimization and the LiDAR point cloud of the interior of the wooden pagoda to obtain the complete point cloud of the wooden pagoda, which makes the deformation monitoring of the pagoda more detailed and comprehensive. After experimental verification, the overall registration accuracy of the Yingxian wooden pagoda reaches 0.006 m. Compared with the scanning point cloud data in 2018, the model is more accurate and complete. By analyzing and comparing the data of the second floor of the wooden pagoda, we knew that the inclination of a second bright layer and a second dark layer is still developing steadily. Overall, the western outer trough inclines thoughtfully, and the column frame slopes from southwest to northeast. Some internal columns showed a negative offset in 2020, and the deformation analysis of a single column was realized by comparing it with the standard column model. The main contribution of this method lies in the effective integration of UAV images and point cloud data to provide accurate data sources for good modeling. This research will provide theoretical and methodological support for the digital protection of architectural heritage and GIS data modeling. The analysis results can provide a scientific basis for the restoration scheme design.

Winter adverse driving dataset for autonomy in inclement winter weather

The availability of public datasets with annotated light detection and ranging (LiDAR) point clouds has advanced autonomous driving tasks, such as semantic and panoptic segmentation. However, there is a lack of datasets focused on inclement weather. Snow and rain degrade visibility and introduce noise in LiDAR point clouds. In this article, summarize a 3-year winter weather data collection effort and introduce the winter adverse driving dataset. It is the first multimodal dataset featuring moderate to severe winter weather—weather that would cause an experienced driver to alter their driving behavior. Our dataset features exclusively events with heavy snowfall and occasional white-out conditions. Data are collected using high-resolution LiDAR, visible as well as near infrared (IR) cameras, a long wave IR camera, forward-facing radio detection and ranging, and Global Navigation Satellite Systems/Inertial Measurement Unit units. Our dataset is unique in the range of sensors and the severity of the conditions observed. It is also one of the only data sets to focus on rural and semi-rural environments. Over 36 TB of adverse winter data have been collected over 3 years. We also provide dense point-wise labels to sequential LiDAR scans collected in severe winter weather. We have labeled and will make available around 1000 sequential LiDAR scenes, amounting to over 7 GB or 3.6 billion labeled points. This is the first point-wise semantically labeled dataset to include falling snow.

Feature- and Distribution-Based LiDAR SLAM With Generalized Feature Representation and Heuristic Nonlinear Optimization

In this article, we propose a novel 3-D light detection and ranging (LiDAR) simultaneous localization and mapping (SLAM) method, which contains feature-based fast scan matching, distribution-based keyframe matching, and loop closure. First, by incorporating feature-based and distribution-based LiDAR scan matching into our SLAM system, we combine the advantages of the two mechanisms and formulate a more efficient and precise front end. Second, considering the local surface reconstructing errors caused by the existing feature representation mechanisms, we propose a generalized representation mechanism for geometry feature representation within the distribution-based framework. Third, due to the existence of noisy point correspondences during the pose estimation optimization process, we propose a novel heuristic knowledge to distinguish high-quality matching error items and put more weights on them during the iterative objective function construction process. Extensive experiments are conducted on public and custom datasets, including KITTI, Mulran, NCLT, Stevens, and USTC-VLP16 datasets collected by our mobile platform. The experimental results demonstrate that the proposed mechanisms can effectively promote the LiDAR SLAM system performance and demonstrate competitive performance compared with state-of-the-art methods. We will make our source code open to serve as a new baseline for the robotic community.

Solar energy digitalization at high latitudes: A model chain combining solar irradiation models, a LiDAR scanner, and high-detail 3D building model

Solar mapping can contribute to exploiting more efficiently the solar energy potential in cities. Solar maps and 3D solar cadasters consist of visualization tools for solar irradiation analysis on urban surfaces (i.e., orography, roofs, and façades). Recent advancements in solar decomposition and transposition modeling and Light Detection and Ranging (LiDAR) scanning enable high levels of detail in 3D solar cadasters, in which the façade domain is considered beside the roof. In this study, a model chain to estimate solar irradiation impinging on surfaces with different orientations at high latitudes is developed and validated against experimental data. The case study is the Zero Emission Building Laboratory in Trondheim (Norway). The main stages of the workflow concern (1) data acquisition, (2) geometry detection, (3) solar radiation modeling, (4) data quality check, and (5) experimental validation. Data are recorded from seven pyranometers installed on the façades (4), roof (2), and pergola (1) and used to validate the Radiance-based numerical model over the period between June 21st and September 21st. This study investigates to which extent high-resolution data sources for both solar radiation and geometry are suitable to estimate global tilted irradiation at high latitudes. In general, the Radiance-based model is found to overestimate solar irradiation. Nonetheless, the hourly solar irradiation modeled for the two pyranometers installed on the roof has been experimentally validated in accordance with ASHRAE Guideline 14. When monthly outcomes are considered for validation, the east and the south pyranometers are validated as well. The achieved results build the ground for the further development of the 3D solar cadaster of Trondheim.

Morphometric constraints on the formation of new terrestrial analogs for planetary pits

The origin of geological depressions abounding on Mars and other planetary bodies remains poorly understood, partially due to the limited variability in the geological settings of existing terrestrial analogs. Here, we present a new terrestrial analog that is located at the northwestern margin of the Levantine volcanic field of Harrat Ash−Shaam along the Dead Sea Transform. The analog site consists of tens of geological depressions (locally named “juba”) that morphologically resemble Martian bowl-shaped pits and occur within a Pleistocene basaltic plateau that overlies Meso-Cenozoic carbonates. To constrain plausible formation mechanisms for the juba depressions, we carried out detailed field mapping and morphometric analyses using a 0.25 m/pixel digital terrain model (DTM) derived from airborne light detection and ranging (LiDAR) survey covering 34 km2 of the study area, and centimeter-scale, ground-based LiDAR scans of selected juba depressions. We show that variable magnitudes of slope asymmetry between north- and south-facing walls within the juba depressions, along with different degrees of sediment infilling, provide effective proxies for the relative geomorphic maturity of these landforms, and in turn indicate asynchronous formation of the juba depressions after the Pleistocene emplacement of the Harrat Ash−Shaam basalts in the study area. Our findings preclude formation of the juba depressions by phreatomagmatic explosions and instead point toward collapse into missing subsurface volume. In a broader context, we propose that the morphometric analyses developed herein to distinguish between plausible juba formation mechanisms in the Harrat Ash−Shaam volcanic field can be extended to better constrain the formation mechanisms of similar pit features on Mars and other planetary bodies.

Metaheuristics for the optimization of Terrestrial LiDAR set-up

3D point clouds have a significant impact on a wide range of applications, although their acquisition is frequently conditioned by the occlusion of the objects in the scene. To address this problem, this paper describes an approach for optimizing LiDAR (Light Detection and Ranging) surveys using metaheuristics such as local searches and genetic algorithms. The method generates a set of optimal scanning locations to densely cover the real-world environment represented through 3D synthetic models. Compared to previous research, this paper handles 3D occlusion by varying the height of the sensor. Also, previously used metrics are compressed into three functions to avoid multi-objective optimization. Regarding performance, a LiDAR scanning solution based on GPU (Graphics Processing Unit) hardware is used. Several tests were conducted to show that the combination of local searches and genetic algorithms generates a reduced set of locations capable of optimizing the scanning of buildings.

Visual and light detection and ranging-based simultaneous localization and mapping for self-driving cars

<span lang="EN-US">In recent years, there has been a strong demand for self-driving cars. For safe navigation, self-driving cars need both precise localization and robust mapping. While global navigation satellite system (GNSS) can be used to locate vehicles, it has some limitations, such as satellite signal absence (tunnels and caves), which restrict its use in urban scenarios. Simultaneous localization and mapping (SLAM) are an excellent solution for identifying a vehicle’s position while at the same time constructing a representation of the environment. SLAM-based visual and light detection and ranging (LIDAR) refer to using cameras and LIDAR as source of external information. This paper presents an implementation of SLAM algorithm for building a map of environment and obtaining car’s trajectory using LIDAR scans. A detailed overview of current visual and LIDAR SLAM approaches has also been provided and discussed. Simulation results referred to LIDAR scans indicate that SLAM is convenient and helpful in localization and mapping.</span>

Water leakage mapping in concrete railway tunnels using LiDAR generated point clouds

Detection of water leakage areas is one of the most important tasks of regular underground tunnel inspections. Current manual visual assessments are plagued by reduced efficiency and poor accuracy, requiring personnel to physically access tunnels to conduct assessments. This form of monitoring is no longer sustainable and thus a more cost-effective method is required. Light detection and ranging (LiDAR) is one such method and commonly known for its ability to generate high-density 3D coordinated point clouds of scanned environments. In addition to point coordinates an intensity value, highly dependent on surface properties and the backscattered energy of the laser beam, is recorded. In this study properties such as surface colour, roughness and state of saturation are reviewed. Results of systematic tests of different colour and concrete targets, scanned using a mobile LiDAR scanner, are discussed. The aim of the research was to prove the aptitude for remote sensing of water leakage areas in underground tunnels and provide an automated workflow to extract quantitative information of each leak. Field tests demonstrated the ability to rapidly capture, identify, extract and record areas of water leakage based on the intensity and spatial information of tunnel point cloud data. This is particularly useful as water ingress is known to degrade concrete, resulting in the earlier onset of corrosion, spalling and loss of strength. The LiDAR scanner used here proved capable of reducing survey time and provided inspectors with a complete 3D model of the tunnel which was supplemented with quantitative leakage information (location and area).

2D FPCB micromirror for scanning LIDAR

This paper presents a 2D flexible printed circuit board (FPCB) micromirror and a scanning 3D light detection and ranging (LIDAR) based on it by integrating the 2D FPCB micromirror with a commercially available single point LIDAR. The 2D FPCB micromirror retains the benefits of previously developed 1D FPCB micromirrors, i.e. large aperture and low cost while providing rotation of the mirror plate about two orthogonal axes to be able to scan a laser beam about both vertical and horizontal axes to achieve 2D scanning. One 2D FPCB micromirror is integrated with a single point LIDAR to achieve a 3D scanning LIDAR, which, in comparison to the previously developed 1D FPCB micromirror based 3D LIDAR, achieved more compact structure and easier fabrication/assembly due to no strict requirement on the alignment between two micromirrors while only one 2D micromirror rather than two 1D micromirrors used. Prototypes of the 2D FPCB micromirror and the 3D LIDAR based on it are fabricated and tested. The test results demonstrate that the 2D FPCB micromirror based 3D LIDAR achieved a volume reduction over the previous 1D FPCB micromirror based 3D LIDAR from 1042 cm3 to 754 cm3 with a field of view of 40°× 24° at 150 Hz horizontal scanning and 2 Hz vertical scanning.

Recognition of ecological vegetation fairy circles in intertidal salt marshes from UAV LiDAR point clouds

• Fairy circles are detected from 3D point clouds of UAV LiDAR for the first time. • A new model is proposed to correct the intensity data of UAV LiDAR. • Merely corrected intensity data can reliably be used for point cloud filtering. • Absent instantaneous UAV LiDAR scanning distances and scan angles are recovered. • The proposed method is promising for spatial patterns recognition in other scenes. Fairy circles (FC) are a type of spatial self-organized patterns that widely exist in various vegetation ecosystems and the accurate detection and quantitative characterization of these mysterious circles remain a technical challenge. In this study, vegetation FC in intertidal salt marshes are recognized from the derived reflectance information (backscattered intensity) and geometric quantities of light detection and ranging (LiDAR) carried on unmanned aerial vehicle (UAV). The specular effect on the UAV LiDAR intensity data over nadir regions of wet salt marshes is eliminated using the laser radar equation and Phong model where the absent distances and incidence angles are approximately retrieved on the basis of geometric and temporal relations in data collection. The FC are progressively recovered through three interconnected procedures. First, the retrieved reflectance information is used to discriminate the mudflat and vegetation points. Second, a spatial connectivity clustering algorithm is utilized on the extracted vegetation points to form individual spatially disconnected clusters. Finally, FC and regular vegetation are successfully recognized by jointly using the salient, size, and circularity features of the generated clusters. A multi-echo UAV LiDAR system is used for data collection at an intertidal salt marsh to assess the feasibility and prospects of the proposed method. Taking the manual detection results from the orthophoto generated by images of a UAV camera system as a reference, the missing detection rate, false detection rate, and area detection error of the proposed method are 6%, 9%, and 5%, respectively. Results suggest that UAV LiDAR is an extremely promising technique to characterize the geometric properties (e.g., location, size, and quantity) of FC from a holistic perspective.

DeLightLCD: A Deep and Lightweight Network for Loop Closure Detection in LiDAR SLAM

Loop closure detection (LCD) is a critical yet still open technique to enhance the performance of simultaneous localization and mapping (SLAM). In this article, a very deep and lightweight neural network deep and lightweight loop closure detection (DeLightLCD) is proposed to enable efficient LCD. The architecture of the network contains two key modules: 1) a Siamese feature extraction module and 2) a dual-attention-based feature difference module. A very deep but lightweight feature extraction network is designed to extract high-dimensional and discriminative features. The dual-attention-based feature difference network generates feature difference maps to identify the difference between a pair of light detection and ranging (LiDAR) scans. Besides, a loop closure candidate rapid selection method is proposed to extract loop closure candidates from the LiDAR scan sequence. The workflow enables real-time LCD in large-scale environments. This approach was evaluated on the open-source KITTI odometry benchmark and Ford campus datasets. The experimental results show that this method outperforms the state-of-the-art methods. Although the model was trained only on the KITTI dataset, it also demonstrated superior performance on the Ford campus dataset. In particular, the proposed approach is more lightweight and highly efficient than the current existing approaches.