Mobile Laser Scanning Point Clouds Research Articles

Road Boundary Extraction Method from Mobile Laser Scanning Point Clouds

Abstract With the rapid development of mobile laser scanning (MLS) technology, high-precision three-dimensional point cloud data has shown great potential in different fields such as topographic mapping, road asset management and smart city construc-tion. Three-dimensional point cloud data contains not only position information, but also the shape and attributes of the target, which is very convenient for obtaining road information. Accurate extraction of the road boundary is a basic task for obtaining road infrastructural data, which can support the generation of high-precision maps, vehicle navigation and auton-omous driving. However, road boundary extraction in urban environments is easily occluded such as vehicles and pedestri-ans on the road, which leads to problems such as difficulty and incomplete extraction of MLS point cloud road boundaries. To address this problem, this study proposes a road boundary extraction method that integrates pavement edge information, accurately considers the position of the road boundary from two dimensions, eliminates false boundaries, and completes the missing boundary through the extracted boundary spatial relationship. First, grid elevation filtering is used to remove high-level non-ground points. Then the pavement edges and curb stone points are extracted from the preprocessed point cloud, and they are superimposed to remove false boundary points to obtain accurate road boundaries. Finally, based on the spatial relationship of road boundaries, missing parts are detected and repaired to obtain complete road boundaries. Experimental results show that the accuracy on real road scenes exceeds 98%, the completeness rate is above 91%, and the extraction quality is above 90%, which verifies the effectiveness and accuracy of this method.

An Individual Tree Detection and Segmentation Method from TLS and MLS Point Clouds Based on Improved Seed Points

Individual Tree Detection and Segmentation (ITDS) is a key step in accurately extracting forest structural parameters from LiDAR (Light Detection and Ranging) data. However, most ITDS algorithms face challenges with over-segmentation, under-segmentation, and the omission of small trees in high-density forests. In this study, we developed a bottom–up framework for ITDS based on seed points. The proposed method is based on density-based spatial clustering of applications with noise (DBSCAN) to initially detect the trunks and filter the clusters by a set threshold. Then, the K-Nearest Neighbor (KNN) algorithm is used to reclassify the non-core clustered point cloud after threshold filtering. Furthermore, the Random Sample Consensus (RANSAC) cylinder fitting algorithm is used to correct the trunk detection results. Finally, we calculate the centroid of the trunk point clouds as seed points to achieve individual tree segmentation (ITS). In this paper, we use terrestrial laser scanning (TLS) data from natural forests in Germany and mobile laser scanning (MLS) data from planted forests in China to explore the effects of seed points on the accuracy of ITS methods; we then evaluate the efficiency of the method from three aspects: trunk detection, overall segmentation and small tree segmentation. We show the following: (1) the proposed method addresses the issues of missing segmentation and misrecognition of DBSCAN in trunk detection. Compared to using DBSCAN directly, recall (r), precision (p), and F-score (F) increased by 6.0%, 6.5%, and 0.07, respectively; (2) seed points significantly improved the accuracy of ITS methods; (3) the proposed ITDS framework achieved overall r, p, and F of 95.2%, 97.4%, and 0.96, respectively. This work demonstrates excellent accuracy in high-density forests and is able to accurately segment small trees under tall trees.

RailPC: A large‐scale railway point cloud semantic segmentation dataset

AbstractSemantic segmentation in the context of 3D point clouds for the railway environment holds a significant economic value, but its development is severely hindered by the lack of suitable and specific datasets. Additionally, the models trained on existing urban road point cloud datasets demonstrate poor generalisation on railway data due to a large domain gap caused by non‐overlapping special/rare categories, for example, rail track, track bed etc. To harness the potential of supervised learning methods in the domain of 3D railway semantic segmentation, we introduce RailPC, a new point cloud benchmark. RailPC provides a large‐scale dataset with rich annotations for semantic segmentation in the railway environment. Notably, RailPC contains twice the number of annotated points compared to the largest available mobile laser scanning (MLS) point cloud dataset and is the first railway‐specific 3D dataset for semantic segmentation. It covers a total of nearly 25 km railway in two different scenes (urban and mountain), with 3 billion points that are finely labelled as 16 most typical classes with respect to railway, and the data acquisition process is completed in China by MLS systems. Through extensive experimentation, we evaluate the performance of advanced scene understanding methods on the annotated dataset and present a synthetic analysis of semantic segmentation results. Based on our findings, we establish some critical challenges towards railway‐scale point cloud semantic segmentation. The dataset is available at https://github.com/NNU‐GISA/GISA‐RailPC, and we will continuously update it based on community feedback.

Evaluating the efficacy of sampling acquisition paths for mapping vegetation structure using terrestrial mobile laser scanning.

Mobile Laser Scanning (MLS) is rapidly increasing in popularity for the capture and quantification of vegetation structure in natural landscapes. However, understanding the optimal implementation for forest capture in the field is still limited, particularly regarding the choice in acquisition path and length followed during data capture. How variable is the characterisation of vegetation structure when using different acquisition paths, and is this difference likely to be significant for most users? In this study we compared four acquisition path designs commonly cited in the literature to determine the importance of this choice for users when capturing vegetation structure. MLS point clouds were systematically captured to repeatedly survey study plots in a closed-canopy forest ecosystem in south-eastern Australia. Digital elevation models, canopy height models, and vertical voxel occupancy profiles were derived to illustrate the sensitivity of the path's length and configuration to variability in the quantification of vegetation structure. We found strong agreement between the acquisition path designs that increased as path length increased. No significant differences were found in digital elevation models derived at 75% and 100% path lengths. However, differences were observed in two and four of the 25% and 50% path length plots respectively. Significant differences in canopy height model data were only found in one plot at the 50% path length. Mean differences to the reference digital elevation models was 0.09 m across all plots, and 0.4 m for the canopy height models. Voxel occupancy profiles showed the greatest agreement between the understory and canopy where the mean difference between acquisition paths across all plots was 5.72%. Our findings have significant implications for the use of MLS from an operational perspective, as they illustrate the variability and reliability of their use in natural forested landscapes. Users can choose to balance their choice in acquisition path design with path length to meet requirements such as efficiency or ease of data capture in the field.

DAAL-WS: A weakly-supervised method integrated with data augmentation and active learning strategies for MLS point cloud semantic segmentation

Mobile laser scanning (MLS) point clouds have increasingly been a significant data source for acquiring accurate three-dimensional (3D) semantic information from complex scenes. However, most current point cloud semantic segmentation methods heavily depend on a huge number of manually labeled training samples, which is labor-intensive and time-consuming. To address the above challenge, a weakly-supervised MLS point cloud semantic segmentation method integrated with data augmentation and active learning strategies is proposed (termed DAAL-WS). By taking advantage of our previously presented multi-branch weakly-supervised network (WSPointNet) in weakly supervisory signals and ensemble predictions, the DAAL-WS integrates WSPointNet with two essential components, i.e., an elevation-calibrated Mix3D (EC-Mix3D) data augmentation strategy and a point-level ensemble prediction-based (PLEP) active learning strategy. Specifically, the EC-Mix3D data augmentation strategy leverages elevation information to calibrate sub-point clouds and generates new contextual scenes by mixing the elevation-calibrated sub-point clouds, thereby augmenting the training point cloud distribution. Designed to reduce labeling redundancy, the PLEP active learning strategy selects the most important labeled points for model training. This strategy first measures the uncertainty for each unlabeled point by ensemble predictions and then employs a feature-distance suppression module to select the significant and discriminating unlabeled points for manual labeling. The proposed DAAL-WS method was evaluated on three public MLS datasets, including Toronto3D, Paris-Lille-3D, and WHU-MLS datasets, on which DAAL-WS obtained a competitive performance over fully-supervised baselines using only 0.015 %, 0.03 %, and 0.03 % labeled points, with mean Intersection over Union (mIoU) scores of 81.91 %, 82.59 %, and 60.36 %, respectively.

Quantification of the spatial distribution of individual mangrove tree species derived from LiDAR point clouds

AbstractMangrove forests have unquestionably high environmental and ecological value. Mangrove trees are believed to have habitat zonation that is controlled mainly by the relative sea level. However, earlier discussions of mangrove habitats have remained limited in terms of their quality and quantity because of a lack of high-resolution spatial information of microtopography and trees. To clarify mangrove habitability over a wide forest area, we compounded mobile laser scanning (MLS) and aerial laser scanning (ALS) LiDAR dataset of the Miyara River mangrove on Ishigaki Island, Okinawa, Japan. The MLS provided sub-canopy data, while the unmanned aerial vehicle ALS data mainly provided a point cloud of the canopy. We corrected point clouds and combined these data. The results indicated that ALS is unable to reconstruct the microtopography of the dense mangrove area well. Moreover, tree species were not identifiable from the ALS data. However, by applying MLS to the mangrove forest, we obtained high-resolution microtopography and tree information inside the forest, although the measurement area was limited to comparison with ALS. By combining ALS and MLS point clouds, 3D point clouds of the forest were well reconstructed. From these point clouds, a high-resolution digital elevation model was created. Subsequently, we segmented trees individually from composite MLS point clouds and identified each tree species. Consequently, the spatial distribution of thousands of mangrove trees was reconstructed at the Miyara River mouth. The spatial distribution of mangrove tree species together with earlier aerial photographs suggests that mangrove species have been segregated in accordance with changes in their elevation and environment over 40 years. Our findings suggest that the distribution of the species changed sensitively along with dynamic variation of the microtopography.

A Rapid Segmentation Method of Highway Surface Point Cloud Data Based on a Supervoxel and Improved Region Growing Algorithm

Mobile laser scanning (MLS) systems have become an important technology for collecting and measuring road information for highway maintenance and reconstruction services. However, the efficient and accurate extraction of unstructured road surfaces from MLS point cloud data collected on highways is challenging. Specifically, the complex and unstructured characteristics of road surveying point cloud data lead to traditional 3D point cloud segmentation. When traditional 3D point cloud algorithms extract unstructured road surfaces, over-segmentation and under-segmentation often occur, which affects efficiency and accuracy. To solve these problems, this study introduces an enhanced road extraction method that integrates supervoxel and trajectory information into a traditional region growing algorithm. The method involves two main steps: first, a supervoxel data structure is applied to reconstruct the original MLS point cloud data, which diminishes the calculation time of the point cloud feature vector and accelerates the merging speed of a similar region; second, the trajectory information of the vehicle is used to optimize the seed selection strategy of the regio growing algorithm, which improves the accuracy of road surface extraction. Finally, two typical highway section tests (flat road and slope road) were conducted to validate the positioning performance of the proposed algorithm in an MLS point cloud. The results show that, compared with three kinds of traditional road surface segmentation algorithms, our method achieves an average extraction recall and precision of 99.1% and 96.0%, and by calculating the recall and precision, an F1 score of 97.5% can be obtained to evaluate the performance of the proposed method, for both datasets. Additionally, our method exhibits an average road surface extraction time that is 45.0%, 50.3%, and 55.8% faster than those of the other three automated segmentation algorithms.

Tree Diameter at Breast Height Extraction Based on Mobile Laser Scanning Point Cloud

The traditional measurement method (e.g., field survey) of tree diameter circumference often has high labor costs and is time-consuming. Mobile laser scanning (MLS) is a powerful tool for measuring forest diameter at breast height (DBH). However, the accuracy of point cloud registration seriously affects the results of DBH measurements. To address this issue, this paper proposes a new method for extracting tree DBH parameters; it achieves the purpose of efficient and accurate extraction of tree DBH by point cloud filtering, single-tree instance segmentation, and least squares circle fitting. Firstly, the point cloud data of the plantation forest samples were obtained by a self-constructed unmanned vehicle-mounted mobile laser scanning system, and the ground point cloud was removed using cloth simulation filtering (CSF). Secondly, fast Euclidean clustering (FEC) was employed to segment the single-tree instances, and the point cloud slices at breast height were extracted based on the point sets of single-tree instances, which were then fitted in two dimensions using the horizontally projected point cloud slices. Finally, a circle fitting algorithm based on intensity weighted least squares (IWLS) was proposed to solve the optimal circle model based on 2D point cloud slices, to minimize the impact of misaligned point clouds on DBH measures. The results showed that the mean absolute error (MAE) of the IWLS method was 2.41 cm, the root mean square error (RMSE) was 2.81 cm, and the relative accuracy was 89.77%. Compared with the random sample consensus (RANSAC) algorithm and ordinary least squares (OLS), the MAE was reduced by 36.45% and 9.14%, the RMSE was reduced by 40.90% and 12.26%, and the relative accuracy was improved by 8.99% and 1.63%, respectively. The R2 value of the fitted curve of the IWLS method was the closest to 1, with the highest goodness of fit and a significant linear correlation with the true value. The proposed intensity weighted least squares circle-fitting DBH extraction method can effectively improve the DBH extraction accuracy of mobile laser scanning point cloud data and reduce the influence of poorly aligned point clouds on DBH fitting.

BEYOND GROUND CONTROL POINTS: COST-EFFECTIVE 3D BUILDING RECONSTRUCTION THROUGH GNSS-INTEGRATED PHOTOGRAMMETRY

Abstract. The process of 3D building modeling serves a multitude of practical and strategic purposes across diverse industries. Building a 3D model involves employing a range of techniques and technologies. Among these, the most used methods include 3D laser scanning and photogrammetry, whether applied at close-range or through the use of Unmanned Aerial Systems (UAS). In photogrammetry, ground control points (GCPs) are generally needed to scale and georeference the digital reconstruction process, but it is a timeconsuming practice or sometimes impractical or dangerous. This paper aims to evaluate the efficiency of two integrated devices to perform photogrammetric 3D reconstruction without GCPs. They are both composed by a Sony ZV1 camera coupled with two different RTK/PPK GNSS system: the Emlid Reach RS2 GNSS receiver and the Emlid Reach M2 module with a multi-band GNSS helical antenna. Different sets of images were acquired with the two proposed devices for the lever-arm estimation and to perform the 3D surveying of the Galata monastery historical monument. The accuracy of the process and derived dense point clouds is assessed by comparing them with GCPs and a reference point cloud derived by fusing an UAS and a high-resolution mobile laser scanning point cloud. The ultimate goal is to obtain a 3D building model without the use of GCPs in the process of bundle block adjustment with centimeter accuracy.

A high-accuracy multi-temporal laser point cloud registration method for providing a detailed expression of the urban environment

In consideration of the significant changes in urban environmental scenes and the impact of factors such as signal obstruction on positioning, the spatial positions of multi-temporal Mobile Laser Scanning (MLS) point clouds collected in the same area are inconsistent, which makes it challenging to provide a complete expression of the urban road environment. To address this issue, a precise registration algorithm based on pole-like targets is proposed. The core steps include registration primitive extraction: a cylindrical spatial neighborhood is established based on the distinctive key points, and precise extraction of pole-like geospatial point cloud data is performed using the Euclidean clustering method with the normal curvature constraint. Achieving optimal spatial transformation: a circle center fitting method with additional Mean Absolute Error (MAE) loss function constraints was developed based on the stable elemental unit which was utilized to accurately construct homonymous feature point pairs. Validation results from multiple real urban point cloud datasets collected from different experimental areas and platforms demonstrate that our method achieves superior registration performance, with mean residuals below 5 cm in both the planar and elevation directions. The proposed method exhibits strong universality and registration robustness for multi-temporal point cloud data of the homologous or cross-sources.

Mobile Laser Scanning Point Cloud Classification Based on Data Augmentation and Mask Learning

Mobile Laser Scanning Point Cloud Classification Based on Data Augmentation and Mask Learning

TOWARDS DETECTING SPEED BUMPS FROM MLS POINT CLOUDS DATA

Abstract. High definition maps contain a high level of information about the road and its features. This includes traffic signs, speed limits and lane markings, that are not generally included on regular maps. However, they may still lack critical information, such as the locations of speed bumps. Without the information about the whereabouts of speed bumps the passengers’ comfort and safety, as well as the condition of the car may be jeopardized. There are currently methods for detecting speed bumps that use changes in acceleration or 3D cameras. However, these approaches are susceptible to external influences interfering with the recorded data. Hence, it is desirable to have this information stored and available through HD maps. In this work, we tested two deep learning-based approaches (PointNet++ and PointCNN) and compared the results with conventional region growing method, in order to find out pros and cons of the modern deep learning-based methods. For our test, we used MLS (mobile laser scanning) point clouds data in Trondheim, Norway.

Weakly supervised semantic segmentation of mobile laser scanning point clouds via category balanced random annotation and deep consistency‐guided self‐distillation mechanism

AbstractScene understanding of mobile laser scanning (MLS) point clouds is vital in autonomous driving and virtual reality. Most existing semantic segmentation methods rely on a large number of accurately labelled points, which is time‐consuming and labour‐intensive. To cope with this issue, this paper explores a weakly supervised learning (WSL) framework for MLS data. Specifically, a category balanced random annotation (CBRA) strategy is employed to obtain balanced labels and enhance model performance. Next, based on KPConv‐Net as a backbone network, a WSL semantic segmentation framework is developed for MLS point clouds via a deep consistency‐guided self‐distillation (DCS) mechanism. The DCS mechanism consists of a deep consistency‐guided self‐distillation branch and an entropy regularisation branch. The self‐distillation branch is designed by constructing an auxiliary network to maintain the consistency of predicted distributions between the auxiliary network and the original network, while the entropy regularisation branch is designed to increase the confidence of the network predicted results. The proposed WSL framework was evaluated on the WHU‐MLS, NPM3D and Toronto3D datasets. By using only 0.1% labelled points, the proposed WSL framework achieved a competitive performance in MLS point cloud semantic segmentation with the mean Intersection over Union (mIoU) scores of 60.08%, 72.0% and 67.42% on the three datasets, respectively.

PatchAugNet: Patch feature augmentation-based heterogeneous point cloud place recognition in large-scale street scenes

Point Cloud Place Recognition (PCPR) in street scenes is an essential task in the fields of autonomous driving, robot navigation, and urban map updating. However, the domain gap between heterogeneous point clouds and the difficulty of feature characterization in large-scale complex street scenes pose significant challenges for existing PCPR methods. Most PCPR methods only take into account point clouds collected by the same platforms and sensors, thus they are with poor domain transferability. In this paper, we propose PatchAugNet, which utilizes patch feature augmentation and adaptive pyramid feature aggregation to achieve better performance and generalizability for Heterogeneous Point Cloud-based Place Recognition (HPCPR) tasks. Firstly, multi-scale local features are extracted by the pyramid feature extraction module. Secondly, local features are enhanced by the patch feature augmentation module to overcome the domain gap problem and achieve better feature representation as well as network generalizability. Finally, a global feature is generated using an adaptive pyramid feature aggregation module, which automatically adjusts and balances the proportion of intra-scale and inter-scale features according to the scene content. To evaluate the performance of PatchAugNet, a large-scale heterogeneous point cloud dataset consisting of high-precision Mobile Laser Scanning (MLS) point clouds and helmet-mounted Portable Laser Scanning (PLS) point clouds is collected. The dataset covers various street scenes with a length of over 20 km. The comprehensive experimental results indicate that PatchAugNet achieves State-Of-The-Art (SOTA) performance with 83.43 % recall@top1% and 60.34 % recall@top1 on unseen large-scale street scenes, outperforming existing SOTA PCPR methods by + 9.57 recall@top1% and + 15.50 recall@top1, while exhibiting better generalizability. For source code and detailed experimental results, please refer to: https://github.com/WHU-USI3DV/PatchAugNet.

Santiago urban dataset SUD: Combination of Handheld and Mobile Laser Scanning point clouds

Santiago Urban Dataset SUD is a real dataset that combines Mobile Laser Scanning (MLS) and Handheld Mobile Laser Scanning (HMLS) point clouds. The data is composed by 2 km of streets, sited in Santiago de Compostela (Spain). Point clouds undergo a manual labelling process supported by both heuristic and Deep Learning methods, resulting in the classification of eight specific classes: road, sidewalk, curb, buildings, vehicles, vegetation, poles, and others. Three PointNet++ models were trained; the first one using MLS point clouds, the second one with HMLS point clouds and the third one with both H&MLS point clouds. In order to ascertain the quality and efficacy of each Deep Learning model, various metrics were employed, including confusion matrices, precision, recall, F1-score, and IoU. The results are consistent with other state-of-the-art works and indicate that SUD is valid for comparing point cloud semantic segmentation works. Furthermore, the survey's extensive coverage and the limited occlusions indicate the potential utility of SUD in urban mobility research.

MuCoGraph: A multi-scale constraint enhanced pose-graph framework for MLS point cloud inconsistency correction

The position consistency of mobile laser scanning (MLS) point clouds is crucial for large-scale applications, and is normally guaranteed by the global navigation satellite system (GNSS) and high-precision inertial measurement unit (IMU) in the data acquisition. However, GNSS-denied environments such as city valleys result in significant position inconsistency for overlapping areas, and it is difficult to automatically locate these inconsistent areas. In this paper, to overcome these problems, we present MuCoGraph, which introduces multi-scale constraints to establish the correct correspondences for revisited areas, and formulates an enhanced pose graph for position inconsistency correction. The georeferenced MLS point cloud is first sliced into segments adaptively, and these segments are then abstracted as graph vertices, which satisfy local geometric consistency and rigid transformation hypotheses. Accurate revisited graph edges are then constructed hierarchically under multi-scale scenery consistency constraints. These revisited edges are initialized based on feature-based correspondence estimation and further unreliable edge pruning. Finally, through combination with virtual co-observations, correspondence-enhanced pose-graph optimization is introduced to globally redistribute the errors and obtain a high-precision point cloud. The proposed method was used to correct the MLS point cloud position inconsistency in three datasets. The average three-dimensional distance of the checkpoints was reduced from 0.362 m, 0.108 m, and 1.027 m to 0.057 m, 0.033 m, and 0.051 m for datasets I, II, and III respectively. In addition, the root-mean-square error of all three datasets was less than 0.04 m after correction. The experiments confirmed that the proposed method can automatically locate and correct the position inconsistency of MLS point clouds, showing good robustness and effectiveness.

Road potholes detection from MLS point clouds

The extraction of pavement damage information is one of the major difficulties in the application research of mobile laser scanning point cloud data. To address the problem of inaccurate detection results by using only relative distance to detect potholes, this paper proposes a novel pothole detection method that combines directed distance and skewed distribution. Firstly, the rapid positioning of the pothole is realized by the directed distance, which is calculated from the points and the local fitted plane. And monomerization and denoising of potential potholes are achieved by density clustering. Then, the new accurate plane is fitted by the surrounding pavement points of the potential pothole to obtain accurate directed distances. The negative skewed distribution of the directed distance histogram and the skewness coefficient are used for the accurate determination of the pothole. Finally, the three-dimensional geometric features of the pothole are extracted. Experiments were carried out on a road with poor road conditions. The experimental results validated the effectiveness and practicality of the proposed method. It can achieve automatic detection of potholes with different shapes and deformation degrees, and has effectively improved the efficiency of automatic road inspection.

Real-time detection of street tree crowns using mobile laser scanning based on pointwise classification

For targeted spraying in urban streets using mobile laser scanning (MLS), it is crucial to continuously detect tree crowns in real time from MLS data collected online. Crown detection based on a pointwise classification method, which predicts whether a point belongs to a crown using a binary classifier composed of a set of local features extracted from the neighbourhood of the point, is suitable for online data processing. However, the computational complexity of pointwise detection methods needs to be reduced to meet the real-time requirements of online targeted spraying. This paper proposes a real-time detection method for street tree crowns using MLS based on pointwise classification. A cubic neighbourhood is adopted from which a set of statistical features are extracted. A coarse-to-fine neighbourhood search method based on an MLS data grid storage structure is proposed to quickly query the neighbours. The classification performance and computational time of single features are evaluated. Then, the features are ranked using the technique for order preference by similarity to the ideal solution (TOPSIS). Finally, a subset of features that are easy to calculate and have high discrimination are selected and fused into a crown detector using a random forest algorithm. During online detection, the point cloud is sampled in terms of the vehicle movement and LiDAR measurement before being detected. An MLS point cloud collected from a 451 m urban street is used in the experiments. When the neighbourhood size is set to 0.3 m, the detection time per scanline is 24.84 ms on the unsampled test set with an F1 score of 0.9927. When the test set is sampled with a total sampling rate of 40, the detection time is reduced to 1% with an F1 score of 0.9906. The experimental results show that the proposed method can provide real-time and accurate detection of street tree crowns for online targeted spraying.

Canopy and surface fuel estimations using RPAS and ground-based point clouds

Abstract Forest management activities intended to reduce wildfire risk rely on accurate characterizations of the amount and arrangement of canopy and surface fuels. Metrics that describe these fuels are typically estimated with various systems that transform plot-level field data into metrics that can be used within fire behaviour models. Remote sensing data have long been used to estimate these metrics across large spatial scales, but more advanced, high-density point clouds have the potential to estimate these metrics with higher accuracy. This study collected LiDAR and digital aerial photogrammetric (DAP) point clouds from a remotely piloted aerial system (RPAS), as well as mobile laser scanning (MLS) point clouds from a mobile ground-based system, and compared their ability to estimate fuel metrics. This involved the extraction of predictor variables from each point cloud, of which small subsets were used to estimate various fuel metrics. These included six overstory canopy metrics (stand height, canopy cover, tree density, canopy fuel load, canopy bulk density and canopy base height), three diameter at breast height (DBH)–related metrics (stand density index, basal area and quadratic mean diameter) and three surface fuel metrics (total woody debris (TWD), coarse woody debris (CWD) and fine woody debris (FWD)). Overall, canopy metrics were estimated most accurately by the RPAS LiDAR models, although none of the point clouds were able to accurately estimate DBH-related metrics. For the other six canopy metrics, RPAS LiDAR models had an average R2 value of 0.70; DAP – 0.63 and MLS – 0.63. CWD (>7 cm) and TWD loads were estimated most accurately by the MLS models (average R2 values – 0.70), followed by the RPAS LiDAR – 0.38 and DAP – 0.13. None of these models were able to accurately estimate FWD loads (≤7 cm in diameter), with the three types of point clouds having a maximum R2 value of 0.08. Overall, this research shows the relative ability of three types of high-density point clouds to estimate metrics relevant for fire behaviour modeling.

D-Net: A Density-Based Convolutional Neural Network for Mobile LiDAR Point Clouds Classification in Urban Areas

The 3D semantic segmentation of a LiDAR point cloud is essential for various complex infrastructure analyses such as roadway monitoring, digital twin, or even smart city development. Different geometric and radiometric descriptors or diverse combinations of point descriptors can extract objects from LiDAR data through classification. However, the irregular structure of the point cloud is a typical descriptor learning problem—how to consider each point and its surroundings in an appropriate structure for descriptor extraction? In recent years, convolutional neural networks (CNNs) have received much attention for automatic segmentation and classification. Previous studies demonstrated deep learning models’ high potential and robust performance for classifying complicated point clouds and permutation invariance. Nevertheless, such algorithms still extract descriptors from independent points without investigating the deep descriptor relationship between the center point and its neighbors. This paper proposes a robust and efficient CNN-based framework named D-Net for automatically classifying a mobile laser scanning (MLS) point cloud in urban areas. Initially, the point cloud is converted into a regular voxelized structure during a preprocessing step. This helps to overcome the challenge of irregularity and inhomogeneity. A density value is assigned to each voxel that describes the point distribution within the voxel’s location. Then, by training the designed CNN classifier, each point will receive the label of its corresponding voxel. The performance of the proposed D-Net method was tested using a point cloud dataset in an urban area. Our results demonstrated a relatively high level of performance with an overall accuracy (OA) of about 98% and precision, recall, and F1 scores of over 92%.