Amount Of Point Cloud Data Research Articles

Massive Point Cloud Processing for Efficient Construction Quality Inspection and Control

The construction of large-scale civil infrastructures requires massive spatiotemporal data to support the management and control of scheduling, quality control, and safety monitoring. Existing artificial-intelligence-based data processing algorithms rely heavily on experienced engineers to adjust the parameters of data processing, which is inefficient and time-consuming when dealing with huge datasets. Limited studies have compared the performance of different algorithms on a unified dataset. This study proposes a framework and evaluation system for comparing different data processing policies for processing huge spatiotemporal data in construction quality control. The proposed method compares the combination of multiple types of algorithms involved in the processing of massive point cloud data. The performance of data processing strategies is evaluated through this framework, and the optimal point cloud processing strategies are explored based on registration accuracy and data fidelity. Results show that a reasonable choice of combinations of point cloud sampling, filtering, and registration algorithms can significantly improve the efficiency of point cloud data processing and satisfy engineering demands for data accuracy and completeness. The proposed method can be applied to the civil engineering problem of processing a large amount of point cloud data and selecting the optimal processing method.

3D Voxel-based Path Planning for AVs in Dynamic Complex Environments

Abstract. With the rapid advancement of science and technology, autonomous driving technology has been widely applied in various fields such as the automotive industry, intelligent warehousing, and emergency rescue. Traditional 2D path planning has limitations in dealing with the complex and dynamic environments of the real world, making it difficult to meet the high precision and complex path planning requirements of autonomous vehicles. With the development of data acquisition technology, large-scale point cloud data has become easily accessible, which can better extract the three-dimensional complex of the environment. Nevertheless, the unstructured nature and vast amount of point cloud data make path planning on it extremely challenging. Voxel models can effectively compress point cloud data and provide neighboring topological structures. In view of this, we propose a voxel-based framework for the path planning of autonomous vehicles in dynamic and complex 3D environments, which involves converting point cloud data into voxels and extracting navigation spaces. Subsequently, deep reinforcement learning techniques are utilized to achieve obstacle avoidance and navigation on the voxel map. It is expected that this framework will contribute to the development of 3D path planning and enhance the utilization of point cloud data in the navigation field.

A novel method for the classification of 3D point clouds based on the improved PointNet++

Abstract In deep learning, point clouds are used as the primary input format for 3D data, which can provide detailed geometric information about objects in the original 3D space. PointNet++ is a deep learning network that uses point cloud data as an input format, which avoids the losses associated with the previous conversion of point cloud into 3D voxelization and a collection of 2D images. Although PointNet++ can directly process point cloud data in various ways, due to the disordered, irregular, and unevenly distributed nature of point cloud data, the effect of extracting point cloud features could be better. The large amount of point cloud data also leads to the training model falling into the local optimal solution, which affects the training results. In recent years, some effective methods and strategies have emerged to address these problems. In this thesis, three methods are proposed based on the PointNet++ network: feature similarity-based attention pooling, small kernel convolution, and diverse branch block method to improve the performance of the PointNet++ network. Experiments show that the improvement methods proposed in this paper effectively improve the feature extraction accuracy, which improves the accuracy of the PointNet++ network for classification on the ModelNet40_Normal_Resampled dataset, with an overall improvement of 1% compared with PointNet++.

Visual reconstruction method of architectural space under laser point cloud big data

In order to solve the problem that the reconstruction accuracy and integrity are affected due to the large amount of point cloud data in the process of building space reconstruction, the visual reconstruction method of building space under laser point cloud big data is studied. The three-dimensional laser scanner is used to collect the laser point cloud big data in the building space, and the laser point cloud big data is organized and processed through three steps: hierarchical calculation of the point cloud pyramid, thinning treatment and block treatment. From the processing results of laser point cloud big data, the line features of building space are extracted based on the improved Mean-shift method, and the continuous broken lines in the point cloud data of building space are extracted by using the double radius threshold line tracing method. According to the feature extraction results of point cloud data in building space, the visual reconstruction of building space is completed through the process of translation matching and space matching. The experimental results show that this method can realize the visual reconstruction of architectural space, and the average reconstruction accuracy is higher than that of 97 %, and the reconstruction completion and smoothness are higher than 95 %.

SPDC: A SUPER-POINT AND POINT COMBINING BASED DUAL-SCALE CONTRASTIVE LEARNING NETWORK FOR POINT CLOUD SEMANTIC SEGMENTATION

Abstract. Semantic segmentation of point clouds is one of the fundamental tasks of point cloud processing and is the basis for other downstream tasks. Deep learning has become the main method to solve point cloud processing. Most existing 3D deep learning models require large amounts of point cloud data to drive them, but annotating the data requires significant time and economic costs. To address the problem of semantic segmentation requiring large amounts of annotated data for training, this paper proposes a Super-point-level and Point-level Dual-scale Contrast learning network (SPDC). To solve the problem that contrastive learning is difficult to train and feature extraction is not sufficient, we introduce super-point maps to assist the network in feature extraction. We use a pre-trained super-point generation network to convert the original point cloud into a super-point map. A dynamic data augmentation(DDA) module is designed for the super-point maps for super-point-level contrastive learning. We map the extracted super-point-level features back to the original point-level scale and conduct secondary contrastive learning with the original point features. The whole feature extraction network is parameter sharing and to reduce the number of parameters we used the lightweight network DGCNN (encoder)+Self-attention as the backbone network. And we did a few-shot pre-training of the backbone network to make the network converge easily. Analogous to CutMix, we designed a new method for point cloud data augmentation called PointObjectMix (POM). This method solves the sample imbalance problem while preserving the overall characteristics of the objects in the scene. We conducted experiments on the S3DIS dataset and obtained 63.3% mIoU. We have also done a large number of ablation experiments to verify the effectiveness of the modules in our method. Experimental results show that our method outperforms the best-unsupervised network available.

Rapid assessment of slope deformation in 3D point cloud considering feature-based simplification and deformed area extraction

Calculating deformation values and determining deformation areas are essential for slope monitoring and safety management. Recently, terrestrial laser scanning has been widely used for deformation monitoring due to to its speed and efficiency. However, handling a large amount of point cloud data to obtain the deformed area is still challenging. To rapidly evaluate slope deformation information, this study proposes a simplification algorithm for point cloud data based on multi-parameter feature preservation. The proposed method retains feature points (i.e. inflection points and edge points) and simplifies data with the octree structure. In addition, a method based on cloud-to-mesh for displacement is used where the winding number is introduced for the signed function. Next, the deformed area over the level of detection is extracted using density-based spatial clustering for applications with a noise clustering algorithm. To verify the reduction method, two types of slope field data are used for experiments. The results reveal that the proposed point cloud approach is superior to the conventional algorithms. Furthermore, the highway slope in Mianyang is selected as a case study to validate the performance of the proposed method. The entire monitoring area is stable with a deformation of approximately 0.43 mm, and only four regions are deformed over the study period. When the mean displacement value is considered in different deformation regions, the minimum deformation is −82.02 mm and the maximum deformation is 85.31 mm. Moreover, comparative experiments on deformation calculation are conducted and reveal the superior performance of the proposed method.

VOXELIZATION TECHNIQUES: DATA SEGMENTATION AND DATA MODELLING FOR 3D BUILDING MODELS

Abstract. Voxelization of data is discretizing the 3D space, in which the simplest form is a single voxel. There is a large number of publications that are related to voxelization. However, this paper focuses on the voxelization technique implemented in 3D building modelling. This paper aims to get the development idea of the voxelization technique throughout these past years to determine the suitable technique and method for including a 3D voxelized building model in Computational Fluid Dynamics (CFD). From the search and analysis, it is found that this technique is not only related to data modelling of the 3D voxelized model; the voxelization technique can also be utilized in the data segmentation process. First, for the data segmentation, the voxelization technique is implemented to manage the large amount of point cloud data that were obtained from the 3D scanner and sensors, which is done by reducing the number of data to avoid data redundancy and unused data using each of the voxels that exist in that environment. Second, for data modelling, popular input data to generate the 3D voxelized model is also in the form of a point cloud. However, there are still other forms, such as line and surface. Nevertheless, this paper reviews the voxelized technique in building modelling despite some data segmentation. The review shows various input data, applications, and techniques associated with the voxelization process based on building model generation. However, there is still room for improvement that allows the 3D model to be modelled in the voxelized form in the CFD domain.

SAUSA: Securing Access, Usage, and Storage of 3D Point CloudData by a Blockchain-Based Authentication Network

The rapid development of three-dimensional (3D) acquisition technology based on 3D sensors provides a large volume of data, which are often represented in the form of point clouds. Point cloud representation can preserve the original geometric information along with associated attributes in a 3D space. Therefore, it has been widely adopted in many scene-understanding-related applications such as virtual reality (VR) and autonomous driving. However, the massive amount of point cloud data aggregated from distributed 3D sensors also poses challenges for secure data collection, management, storage, and sharing. Thanks to the characteristics of decentralization and security, Blockchain has great potential to improve point cloud services and enhance security and privacy preservation. Inspired by the rationales behind the software-defined network (SDN) technology, this paper envisions SAUSA, a Blockchain-based authentication network that is capable of recording, tracking, and auditing the access, usage, and storage of 3D point cloud datasets in their life-cycle in a decentralized manner. SAUSA adopts an SDN-inspired point cloud service architecture, which allows for efficient data processing and delivery to satisfy diverse quality-of-service (QoS) requirements. A Blockchain-based authentication framework is proposed to ensure security and privacy preservation in point cloud data acquisition, storage, and analytics. Leveraging smart contracts for digitizing access control policies and point cloud data on the Blockchain, data owners have full control of their 3D sensors and point clouds. In addition, anyone can verify the authenticity and integrity of point clouds in use without relying on a third party. Moreover, SAUSA integrates a decentralized storage platform to store encrypted point clouds while recording references of raw data on the distributed ledger. Such a hybrid on-chain and off-chain storage strategy not only improves robustness and availability, but also ensures privacy preservation for sensitive information in point cloud applications. A proof-of-concept prototype is implemented and tested on a physical network. The experimental evaluation validates the feasibility and effectiveness of the proposed SAUSA solution.

Automatic Identification and Intelligent Optimization of Tunnel-Free Curve Reconfiguration

Extracting complete cross-sectional geometric features from the large amount of point cloud data acquired by laser scanners plays an important part in the detection of deformations in tunnel inspection projects. Tunnel cross-sections have symmetrical geometric features, and information is traditionally collected manually. The traditional manual extraction of point clouds is inefficient and limited by the subjectivity of the operators when addressing the problems. This paper proposes a new algorithm for the automatic identification of tunnel lining section curves, the rapid separation of common interference targets, and the optimization of curve geometry features. The innovation of this approach lies in the combination of B-spline and Euclidean clustering methods and the comprehensive evaluation of the denoising results in terms of precision, recall, F-score, and rand index (RI). In this way, the automatically extracted health point cloud data are refitted to optimize the tunnel profile model.

A Novel Coding Scheme for Large-Scale Point Cloud Sequences Based on Clustering and Registration

Due to the huge volume of point cloud data, storing and transmitting it is currently difficult and expensive in autonomous driving. Learning from the high-efficiency video coding (HEVC) framework, we propose a novel compression scheme for large-scale point cloud sequences, in which several techniques have been developed to remove the spatial and temporal redundancy. The proposed strategy consists mainly of three parts: intracoding, intercoding, and residual data coding. For intracoding, inspired by the depth modeling modes (DMMs), in 3-D HEVC (3-D-HEVC), a cluster-based prediction method is proposed to remove the spatial redundancy. For intercoding, a point cloud registration algorithm is utilized to transform two adjacent point clouds into the same coordinate system. By calculating the residual map of their corresponding depth image, the temporal redundancy can be removed. Finally, the residual data are compressed either by lossless or lossy methods. Our approach can deal with multiple types of point cloud data, from simple to more complex. The lossless method can compress the point cloud data to 3.63% of its original size by intracoding and 2.99% by intercoding without distance distortion. Experiments on the KITTI dataset also demonstrate that our method yields better performance compared with recent well-known methods. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This article deals with the problem of efficient compression of point cloud sequences that come from light detection and ranging (LiDARs) mounted on autonomous mobile robots. The vast amount of point cloud data could be an important bottleneck for transmission and storage. Inspired by the HEVC algorithm, we develop a novel coding architecture for the point cloud sequence. The scans are divided into intraframe and interframe, which are encoded separately using different techniques. Our method can be used for the compression of LiDAR point cloud sequences or dense LiDAR point cloud map and will significantly reduce the transmission bandwidth and storage spaces. We have to admit that although our method is less effective for real-time solutions, it can be highly efficient for off-line applications. Future studies will concentrate on further optimizing the coding algorithm to reduce the computational complexity and trying to find a balance between them.

MBGRLp: Multiscale Bootstrap Graph Representation Learning on Pointcloud (Student Abstract)

Point cloud has gained a lot of attention with the availability of a large amount of point cloud data and increasing applications like city planning and self-driving cars. However, current methods, often rely on labeled information and costly processing, such as converting point cloud to voxel. We propose a self-supervised learning approach to tackle these problems, combating labelling and additional memory cost issues. Our proposed method achieves results comparable to supervised and unsupervised baselines on the widely used benchmark datasets for self-supervised point cloud classification like ShapeNet, ModelNet10/40.

Early detection of tree encroachment in high voltage powerline corridor using growth model and UAV-borne LiDAR

UAV-borne LiDAR is an innovative and effective technique for tree encroachment detection in high voltage powerline corridor. However, the periodical inspection of the whole powerline corridor is inefficient, as the powerline segments occurring the tree encroachment only account for a very small part. In this paper, taking one segment of the powerline corridor in Taining County, Fujian province, China as the test site, we acquired the point cloud data using a UAV-borne LiDAR, and then combined the tree growth model and two-phase encroachment detection algorithm to realize efficiently early detection of tree encroachment. First, the points of powerlines and trees were classified from the point cloud, and then the individual tree heights and the belonging tree species were extracted. Based on tree plot data, the relationships between tree heights and tree ages were established using Richards growth model. Secondly, the individual tree heights were predicted at given time points, and the tree encroachments were detected in advance according to the required safe distance between powerlines and trees. To tackle the huge amount of point cloud data and calculation, the two-phase tree encroachment detection algorithm based on bounding boxes was applied to replace the traditional point traversal algorithm. As a result, the exact locations of tree encroachment were early detected and the specific encroaching trees were also pre-identified. Lastly, the pre-detected tree encroachment should be verified through field survey, and treated accordingly. Thus, the inspection efficiency would be greatly improved. In accuracy assessment, the coefficient of determination (R2) and the root mean square error (RMSE) of fitted growth model for Masson pine were 0.812 and 2.308 m, and 0.861 and 2.556 m for Eucalyptus, respectively. Compared with point traversal algorithm, the calculation efficiency of the two-phase tree encroachment detection algorithm was improved by nearly 76 times on average.

Towards Human in the Loop Analysis of Complex Point Clouds: Advanced Visualizations, Quantifications, and Communication Features in Virtual Reality.

Multiple fields in biological and medical research produce large amounts of point cloud data with high dimensionality and complexity. In addition, a large set of experiments generate point clouds, including segmented medical data or single-molecule localization microscopy. In the latter, individual molecules are observed within their natural cellular environment. Analyzing this type of experimental data is a complex task and presents unique challenges, where providing extra physical dimensions for visualization and analysis could be beneficial. Furthermore, whether highly noisy data comes from single-molecule recordings or segmented medical data, the necessity to guide analysis with user intervention creates both an ergonomic challenge to facilitate this interaction and a computational challenge to provide fluid interactions as information is being processed. Several applications, including our software DIVA for image stack and our platform Genuage for point clouds, have leveraged Virtual Reality (VR) to visualize and interact with data in 3D. While the visualization aspects can be made compatible with different types of data, quantifications, on the other hand, are far from being standard. In addition, complex analysis can require significant computational resources, making the real-time VR experience uncomfortable. Moreover, visualization software is mainly designed to represent a set of data points but lacks flexibility in manipulating and analyzing the data. This paper introduces new libraries to enhance the interaction and human-in-the-loop analysis of point cloud data in virtual reality and integrate them into the open-source platform Genuage. We first detail a new toolbox of communication tools that enhance user experience and improve flexibility. Then, we introduce a mapping toolbox allowing the representation of physical properties in space overlaid on a 3D mesh while maintaining a point cloud dedicated shader. We introduce later a new and programmable video capture tool in VR and desktop modes for intuitive data dissemination. Finally, we highlight the protocols that allow simultaneous analysis and fluid manipulation of data with a high refresh rate. We illustrate this principle by performing real-time inference of random walk properties of recorded trajectories with a pre-trained Graph Neural Network running in Python.

A novel fast classification filtering algorithm for LiDAR point clouds based on small grid density clustering

Clustering filtering is usually a practical method for light detection and ranging (LiDAR) point clouds filtering according to their characteristic attributes. However, the amount of point cloud data is extremely large in practice, making it impossible to cluster point clouds data directly, and the filtering error is also too large. Moreover, many existing filtering algorithms have poor classification results in discontinuous terrain. This article proposes a new fast classification filtering algorithm based on density clustering, which can solve the problem of point clouds classification in discontinuous terrain. Based on the spatial density of LiDAR point clouds, also the features of the ground object point clouds and the terrain point clouds, the point clouds are clustered firstly by their elevations, and then the plane point clouds are selected. Thus the number of samples and feature dimensions of data are reduced. Using the DBSCAN clustering filtering method, the original point clouds are finally divided into noise point clouds, ground object point clouds, and terrain point clouds. The experiment uses 15 sets of data samples provided by the International Society for Photogrammetry and Remote Sensing (ISPRS), and the results of the proposed algorithm are compared with the other eight classical filtering algorithms. Quantitative and qualitative analysis shows that the proposed algorithm has good applicability in urban areas and rural areas, and is significantly better than other classic filtering algorithms in discontinuous terrain, with a total error of about 10%. The results show that the proposed method is feasible and can be used in different terrains.

Classifying rock slope materials in photogrammetric point clouds using robust color and geometric features

Photogrammetry is increasingly being used to characterize rock slope hazards in mountainous environments. With growth in the amount of point cloud data being collected, there is a need for algorithmic and rapid point cloud interpretation methods to prepare the data for engineering analysis. However, there is a lack of semantic segmentation literature focused on geological datasets and the potential challenges they pose in a practical setting, such as non-ideal data collection parameters, variable lighting conditions, and label noise. This study presents smartphone and UAV photogrammetry datasets including multiple slope morphologies and lighting conditions, manually labels them to create a training database, and classifies points into five geologically relevant categories using Random Forest. Datasets were collected at multiple times of day and seasons to include varying proportions of shadowed areas, examples of clear and overcast skies, and snow. We compared 12 different point cloud feature sets from previous studies, including combinations of geometric, slope, absolute color, and texture features. In addition, we propose multi-scale color standard deviation and the Grey-Level Co-occurrence Matrix as potentially useful descriptors of texture. Accuracy results generally indicate that feature sets containing absolute color features are highly sensitive to changes in lighting conditions and they are not able to discern between various geological materials (soil, talus, rock). In contrast, our novel feature sets with only geometry, slope, and texture were significantly less influenced by lighting and were on average the highest overall accuracy of any feature set tested, with an average overall accuracy of around 80%.

Iterative automatic global registration algorithm for multi-view point cloud of underground tunnel space

Aiming at the narrow and long tunnel structure, few internal features, and a large amount of point cloud data, the existing registration algorithms and commercial software registration results are not ideal, an iterative global registration algorithm is proposed for massive underground tunnel point cloud registration, which is composed of local initial pose acquisition and global adjustment. Firstly, the feature point coordinates in the point cloud are extracted, and then the station-by-station registration is performed according to the Rodrigues matrix. Finally, the registration result is considered as the initial value of the parameter, and the global adjustment of all observations is carried out. The observation values are weighted by the selection weight iteration method and the weights are constantly modified in the iteration process until the threshold conditions are met and the iteration stops. In this paper, the experimental data, made up of 85 stations of point cloud data, are from the Xiamen subway tunnel, which is about 1300 m long. When the accumulated error of station-to-station registration is too large, several stations are regarded as partial wholes, and the optimal registration is achieved through multiple global adjustments, and the registration accuracy is within 5 mm. Experimental results confirm the feasibility and effectiveness of the algorithm, which provides a new method for point cloud registration of underground space tunnel.

3D Fast Object Detection Based on Discriminant Images and Dynamic Distance Threshold Clustering.

The object detection algorithm based on vehicle-mounted lidar is a key component of the perception system on autonomous vehicles. It can provide high-precision and highly robust obstacle information for the safe driving of autonomous vehicles. However, most algorithms are often based on a large amount of point cloud data, which makes real-time detection difficult. To solve this problem, this paper proposes a 3D fast object detection method based on three main steps: First, the ground segmentation by discriminant image (GSDI) method is used to convert point cloud data into discriminant images for ground points segmentation, which avoids the direct computing of the point cloud data and improves the efficiency of ground points segmentation. Second, the image detector is used to generate the region of interest of the three-dimensional object, which effectively narrows the search range. Finally, the dynamic distance threshold clustering (DDTC) method is designed for different density of the point cloud data, which improves the detection effect of long-distance objects and avoids the over-segmentation phenomenon generated by the traditional algorithm. Experiments have showed that this algorithm can meet the real-time requirements of autonomous driving while maintaining high accuracy.

LiDAR Point Cloud Recognition and Visualization with Deep Learning for Overhead Contact Inspection.

As overhead contact (OC) is an essential part of power supply systems in high-speed railways, it is necessary to regularly inspect and repair abnormal OC components. Relative to manual inspection, applying LiDAR (light detection and ranging) to OC inspection can improve efficiency, accuracy, and safety, but it faces challenges to efficiently and effectively segment LiDAR point cloud data and identify catenary components. Recent deep learning-based recognition methods are rarely employed to recognize OC components, because they have high computational complexity, while their accuracy needs to be improved. To track these problems, we first propose a lightweight model, RobotNet, with depthwise and pointwise convolutions and an attention module to recognize the point cloud. Second, we optimize RobotNet to accelerate its recognition speed on embedded devices using an existing compilation tool. Third, we design software to facilitate the visualization of point cloud data. Our software can not only display a large amount of point cloud data, but also visualize the details of OC components. Extensive experiments demonstrate that RobotNet recognizes OC components more accurately and efficiently than others. The inference speed of the optimized RobotNet increases by an order of magnitude. RobotNet has lower computational complexity than other studies. The visualization results also show that our recognition method is effective.

Prioritized Transmission Control of Point Cloud Data Obtained by LIDAR Devices

Smart monitoring, particularly at intersections, is a promising service that is being considered for the concept of smart cities. A network of light detection and ranging (LIDAR) sensors, which generates point cloud data in real time, can be used to detect people's mobility in smart monitoring. Due to the sheer volume of point cloud data, data transmission requires a significant amount of communication resources. In order to monitor people's mobility in real time, it is necessary to reduce the amount of transmission data to shorten delay. Point cloud compression is one method for reducing the amount of data. However, prior works addressing point cloud compression mainly focused on accuracy for the compression of an entire point cloud without considering its spatial characteristics. The more dynamically a spatial region changes, the more important it is when detecting moving objects such as cars, trucks, pedestrians, and bikes in smart monitoring. This paper proposes a prioritized transmission scheme that applies multiple point cloud compression methods to point cloud data according to the spatial importance of the data, i.e., how dynamically spatial regions change. This paper assumes data transmission of point cloud data from multiple LIDAR devices to an edge server and addresses the intra-frame geometry compression of point cloud data. The proposed scheme splits the point cloud into multiple classes according to the spatial importance and applies multiple point cloud compression methods to each class. A numerical study using a real point cloud dataset obtained at an intersection demonstrates the dependencies of quality, volume, and processing time on possible compression format options. The results verify that the proposed scheme reduces the amount of point cloud data drastically while satisfying the quality and processing time requirements.

Research on a Single-Tree Point Cloud Segmentation Method Based on UAV Tilt Photography and Deep Learning Algorithm

Developing a robust point cloud segmentation algorithm for individual trees from an amount of point cloud data has great significance for tracking tree changes. This method can measure the size, growth, and mortality of individual trees to track and understand forest carbon storage and variation. Traditional measurement methods are not only slow but also tardy. In order to obtain forest information better and faster, this article focuses on two aspects: The first is using UAVs to obtain multiview remote sensing images of the forest, and then using the structure from motion algorithm to construct the forest sparse point cloud and patch-based MVS algorithm to construct the dense point cloud. The second is that a targeted point cloud deep learning method is proposed to extract the point cloud of a single tree. The research results show that the accuracy of single-tree point cloud segmentation of deep learning methods is more than 90%, and the accuracy is far better than traditional planar image segmentation and point cloud segmentation. The combination of point cloud data acquisition with UAV remote sensing and point cloud deep learning algorithms can meet the needs of forestry surveys. It is undeniable that this method, as a forestry survey tool, has a large space for promotion and possible future development.