Point Cloud Segmentation Algorithm Research Articles

Point Cloud Measurement of Rubber Tread Dimension Based on RGB-Depth Camera

To achieve an accurate measurement of tread size after fixed-length cutting, this paper proposes a point-cloud-based tread size measurement method. Firstly, a mathematical model of corner points and a reprojection error is established, and the optimal solution of the number of corner points is determined by the non-dominated sorting genetic algorithm II (NSGA-II), which reduces the reprojection error of the RGB-D camera. Secondly, to address the problem of the low accuracy of the traditional pixel metric ratio measurement method, the random sampling consensus point cloud segmentation algorithm (RANSAC) and the oriented bounding box (OBB) collision detection algorithm are introduced to complete the accurate detection of the tread size. By comparing the absolute error and relative error data of several groups of experiments, the accuracy of the detection method in this paper reaches 1 mm, and the measurement deviation is between 0.14% and 2.67%, which is in line with the highest accuracy standard of the national standard. In summary, the RGB-D visual inspection method constructed in this paper has the characteristics of low cost and high inspection accuracy, which is a potential solution to enhance the pickup guidance of tread size measurement.

Extraction of Moso Bamboo Parameters Based on the Combination of ALS and TLS Point Cloud Data.

Extracting moso bamboo parameters from single-source point cloud data has limitations. In this article, a new approach for extracting moso bamboo parameters using airborne laser scanning (ALS) and terrestrial laser scanning (TLS) point cloud data is proposed. Using the field-surveyed coordinates of plot corner points and the Iterative Closest Point (ICP) algorithm, the ALS and TLS point clouds were aligned. Considering the difference in point distribution of ALS, TLS, and the merged point cloud, individual bamboo plants were segmented from the ALS point cloud using the point cloud segmentation (PCS) algorithm, and individual bamboo plants were segmented from the TLS and the merged point cloud using the comparative shortest-path (CSP) method. The cylinder fitting method was used to estimate the diameter at breast height (DBH) of the segmented bamboo plants. The accuracy was calculated by comparing the bamboo parameter values extracted by the above methods with reference data in three sample plots. The comparison results showed that by using the merged data, the detection rate of moso bamboo plants could reach up to 97.30%; the R2 of the estimated bamboo height was increased to above 0.96, and the root mean square error (RMSE) decreased from 1.14 m at most to a range of 0.35-0.48 m, while the R2 of the DBH fit was increased to a range of 0.97-0.99, and the RMSE decreased from 0.004 m at most to a range of 0.001-0.003 m. The accuracy of moso bamboo parameter extraction was significantly improved by using the merged point cloud data.

3D investigation of gully headcut processes: A slicing segmentation based on filed scouring experiments and laser scanning

Gully headcut erosion is a major gully development process contributing to soil-related economic and environmental problems. However, the current understanding of gully headcut is severely lacking due to the difficulty of effectively monitoring relevant processes. In this study, three small-scale experimental plots were established on a field slope of the hilly-gully region of the Chinese Loess Plateau. Fifteen scouring experiments were conducted on each of the plots, while terrestrial laser scanning was used to acquire morphological data of gully headcuts. Using the volume calculations algorithm based on multiscale model-to-model cloud comparison (3D-M3C2), the spatiotemporal distribution of gully headcut erosion was derived, while the erosion mass, deposition mass and sediment yield were also quantified. An improved slicing algorithm and a point cloud segmentation algorithm were proposed to extract the sub-processes of gully headcut erosion. Results showed: (i) Gully headcut erosion was dominated by erosion in cave areas (contributed 67.09% of the sediment yield) and supported by erosion in collapse areas (contributed 33.35% of the sediment yield). This indicated that runoff shear was the main driver of gully headcut erosion and mass movement played a secondary role. (ii) From a vertical perspective, gully headcut erosion was a cyclic process in which erosion of cave areas and collapse areas reinforced each other. From a horizontal perspective, runoff tended to erode the area with weak concavity. (iii) The channel areas contributed the least (even negatively) to sediment yield of gully headcut erosion (−0.44%), while it contained the most complex processes, involving erosion and deposition caused by plunging pool formation and filling, as well as deposition resulting from mass movement and subsequent re-erosion processes. The study provided a valuable reference for the monitoring and process understanding of gully headcut erosion.

A Point Cloud Segmentation Method for Pigs from Complex Point Cloud Environments Based on the Improved PointNet++

In animal husbandry applications, segmenting live pigs in complex farming environments faces many challenges, such as when pigs lick railings and defecate within the acquisition environment. The pig’s behavior makes point cloud segmentation more complex because dynamic animal behaviors and environmental changes must be considered. This further requires point cloud segmentation algorithms to improve the feature capture capability. In order to tackle the challenges associated with accurately segmenting point cloud data collected in complex real-world scenarios, such as pig occlusion and posture changes, this study utilizes PointNet++. The SoftPool pooling method is employed to implement a PointNet++ model that can achieve accurate point cloud segmentation for live pigs in complex environments. Firstly, the PointNet++ model is modified to make it more suitable for pigs by adjusting its parameters related to feature extraction and sensory fields. Then, the model’s ability to capture the details of point cloud features is further improved by using SoftPool as the point cloud feature pooling method. Finally, registration, filtering, and extraction are used to preprocess the point clouds before integrating them into a dataset for manual annotation. The improved PointNet++ model’s segmentation ability was validated and redefined with the pig point cloud dataset. Through experiments, it was shown that the improved model has better learning ability across 529 pig point cloud data sets. The optimal mean Intersection over Union (mIoU) was recorded at 96.52% and the accuracy at 98.33%. This study has achieved the automatic segmentation of highly overlapping pigs and pen point clouds. This advancement enables future animal husbandry applications, such as estimating body weight and size based on 3D point clouds.

Concrete Armour Unit Breakwater Physical Model Monitoring With 3D Modeling Tools

SEABIM® is a patented scan to BIM process that can generate a reliable and complete 3D model of a rubble-mound breakwater precast protective layer (Xbloc®, Accropode™, Core-loc™ etc.). Using computer vision techniques, the known 3D shape of the Concrete Armour Units (CAU) is detected in a high-resolution point cloud, which allows to obtain the position and orientation of each unit relative to the cloud. By superimposing 3D models produced from different scans at different dates, the movement of each block can then be quantified and represented as a vector. For natural riprap armours, a point cloud segmentation algorithm is applied. Each rock is identified, and then geometric characteristics of the armour can be computed (apparent rock diameters, placement density), and the movement between each scan can be monitored. This tool has been applied to numerous real-scale projects since 2019, both in the monitoring of newly built infrastructures and in the asset management phase (Aberdeen South Harbour, Calais port 2015, Nouvelle Route du Littoral etc...). Breakwater physical models in laboratories, in wave flumes or basins, also require monitoring of the movement of reduced-scale blocks between each wave series. With SEABIM® this movement can be computed with accuracy for each CAU, allowing for a consistent evaluation of the armour layer response to wave loads. The automatic block detection process can be launched on the point cloud from a survey of the initial physical model, obtained either by photogrammetry or LiDAR. Then a new survey is done after each wave series in order to create a sequence of 3D models. Those models can be compared by computing the displacement vectors of the centers of gravity of the units. The vectors can be visualized using color graded arrows within the 3D model. All the information can then be exported to a spreadsheet for further analysis. Using this technique, settlement, rocking and sliding effects can be identified. In this paper we will discuss in detail the adaptation of the survey techniques from real scale to laboratory projects (difference in scanning technique, scale effect on point cloud density etc.). Then we will explain the 3D modeling process and the applications of the 3D models for engineering analysis purposes.

Tree parameter extraction in Fokienia hodginsii plantation based on airborne LiDAR data.

Accurate and efficient extraction of tree parameters from plantations lay foundation for estimating individual wood volume and stand stocking. In this study, we proposed a method of extracting high-precision tree parameters based on airborne LiDAR data. The main process included data pre-processing, ground filtering, individual tree segmentation, and parameter extraction. We collected high-density airborne point cloud data from the large-diameter timber of Fokienia hodginsii plantation in Guanzhuang State Forestry Farm, Shaxian County, Fujian Province, and pre-processed the point cloud data by denoising, resampling and normalization. The vegetation point clouds and ground point clouds were separated by the Cloth Simulation Filter (CSF). The former data were interpolated using the Delaunay triangulation mesh method to generate a digital surface model (DSM), while the latter data were interpolated using the Inverse Distance Weighted to generate a digital elevation model (DEM). After that, we obtained the canopy height model (CHM) through the difference operation between the two, and analyzed the CHM with varying resolutions by the watershed algorithm on the accuracy of individual tree segmentation and parameter extraction. We used the point cloud distance clustering algorithm to segment the normalized vegetation point cloud into individual trees, and analyzed the effects of different distance thresholds on the accuracy of indivi-dual tree segmentation and parameter extraction. The results showed that the watershed algorithm for extracting tree height of 0.3 m resolution CHM had highest comprehensive evaluation index of 91.1% for individual tree segmentation and superior accuracy with R2 of 0.967 and RMSE of 0.890 m. When the spacing threshold of the point cloud segmentation algorithm was the average crown diameter, the highest comprehensive evaluation index of 91.3% for individual tree segmentation, the extraction accuracy of the crown diameter was superior, with R2 of 0.937 and RMSE of 0.418 m. Tree height, crown diameter, tree density, and spatial distribution of trees were estimated. There were 5994 F. hodginsii, with an average tree height of 16.63 m and crown diameter of 3.98 m. Trees with height of 15-20 m were the most numerous (a total of 2661), followed by those between 10-15 m. This method of forest parameter extraction was useful for monitoring and managing plantations.

Integrated approach for facility management of existing buildings using point cloud segmentation

PurposeFacility management gained profound importance due to the increasing complexity of different systems and the cost of operation and maintenance. However, due to the increasing complexity of different systems, facility managers may suffer from a lack of information. The purpose of this paper is to propose a new facility management approach that links segmented assets to the vital data required for managing facilities.Design/methodology/approachAutomatic point cloud segmentation is one of the most crucial processes required for modelling building facilities. In this research, laser scanning is used for point cloud acquisition. The research utilises region growing algorithm, colour-based region-growing algorithm and Euclidean cluster algorithm.FindingsA case study is worked out to test the accuracy of the considered point cloud segmentation algorithms utilising metrics precision, recall and F-score. The results indicate that Euclidean cluster extraction and region growing algorithm revealed high accuracy for segmentation.Originality/valueThe research presents a comparative approach for selecting the most appropriate segmentation approach required for accurate modelling. As such, the segmented assets can be linked easily with the data required for facility management.

Improved Tree Segmentation Algorithm Based on Backpack-LiDAR Point Cloud

For extracting tree structural data from LiDAR point clouds, individual tree segmentation is of great significance. Most individual tree segmentation algorithms miss segmentation and misrecognition, requiring manual post-processing. This study utilized a hierarchical approach known as segmentation based on hierarchical strategy (SHS) to improve individual tree segmentation. The tree point cloud was divided into the trunk layer and the canopy layer to carry out trunk detection and canopy segmentation, respectively. The effectiveness of SHS was evaluated on three mixed broadleaf forest plots. The segmentation efficacy of SHS was evaluated on three mixed broadleaf forest plots and compared with the point cloud segmentation algorithm (PCS) and the comparative shortest-path algorithm (CSP). In the three plots, SHS correctly identified all the trunk portion, had a recall (r) of 1, 0.98, and 1, a precision (p) of 1, and an overall segmentation rate (F) of 1, 0.99, and 1. CSP and PCS are less accurate than SHS. In terms of overall plots, SHS had 10%–15% higher F-scores than PCS and CSP. SHS extracted crown diameters with R2s of 0.91, 0.93, and 0.89 and RMSEs of 0.24 m, 0.23 m, and 0.30 m, outperforming CSP and PCS. Afterwards, we evaluate the three algorithms’ findings, examine the SHS algorithm’s parameters and constraints, and discuss the future directions of this research. This work offers an enhanced SHS that improves upon earlier research, addressing missed segmentation and misrecognition issues. It improves segmentation accuracy, individual tree segmentation, and provides both theoretical and data support for the LiDAR application in forest detection.

Point Cloud Segmentation Algorithm Based on Density Awareness and Self-Attention Mechanism

Point Cloud Segmentation Algorithm Based on Density Awareness and Self-Attention Mechanism

Point Cloud Segmentation Algorithm for Vehicle Body Components Based on Principle of Density-Proportional Growth Consistency

Point Cloud Segmentation Algorithm for Vehicle Body Components Based on Principle of Density-Proportional Growth Consistency

A Point Cloud Segmentation Method for Dim and Cluttered Underground Tunnel Scenes Based on the Segment Anything Model

In recent years, point cloud segmentation technology has increasingly played a pivotal role in tunnel construction and maintenance. Currently, traditional methods for segmenting point clouds in tunnel scenes often rely on a multitude of attribute information, including spatial distribution, color, normal vectors, intensity, and density. However, the underground tunnel scenes show greater complexity than road tunnel scenes, such as dim light, indistinct boundaries of tunnel walls, and disordered pipelines. Furthermore, issues pertaining to data quality, such as the lack of color information and insufficient annotated data, contribute to the subpar performance of conventional point cloud segmentation algorithms. To address this issue, a 3D point cloud segmentation framework specifically for underground tunnels is proposed based on the Segment Anything Model (SAM). This framework effectively leverages the generalization capability of the visual foundation model to automatically adapt to various scenes and perform efficient segmentation of tunnel point clouds. Specifically, the tunnel is first sliced along its direction on the tunnel line. Then, each sliced point cloud is projected onto a two-dimensional plane. Various projection methods and point cloud coloring techniques are employed to enhance SAM’s segmentation performance in images. Finally, the semantic segmentation of the entire underground tunnel is achieved by a small set of manually annotated semantic labels used as prompts in a progressive and recursive manner. The key feature of this method lies in its independence from model training, as it directly and efficiently addresses tunnel point cloud segmentation challenges by capitalizing on the generalization capability of foundation model. Comparative experiments against classical region growing algorithms and PointNet++ deep learning algorithms demonstrate the superior performance of our proposed algorithm.

Three-Dimensional Point Cloud Segmentation Algorithm Based on Depth Camera for Large Size Model Point Cloud Unsupervised Class Segmentation.

This paper proposes a 3D point cloud segmentation algorithm based on a depth camera for large-scale model point cloud unsupervised class segmentation. The algorithm utilizes depth information obtained from a depth camera and a voxelization technique to reduce the size of the point cloud, and then uses clustering methods to segment the voxels based on their density and distance to the camera. Experimental results show that the proposed algorithm achieves high segmentation accuracy and fast segmentation speed on various large-scale model point clouds. Compared with recent similar works, the algorithm demonstrates superior performance in terms of accuracy metrics, with an average Intersection over Union (IoU) of 90.2% on our own benchmark dataset.

Segmentation Clustering Algorithm for Point Cloud Processing of Targets on The Ground

After the ground extraction algorithm, there are multiple objects, including the target in the non-ground point cloud. To effectively extract the features of individual objects and overcome the dependence of the traditional algorithm based on the normal vector and curvature of the region growing point cloud segmentation method on the initial point, the normal vector angle in the neighborhood, and the curvature threshold selection. In this paper, we propose a fast clustering segmentation method for region-growing point clouds based on neighborhood density. The test results show that the point cloud segmentation algorithm proposed in this paper has better performance than the traditional Euclidean clustering point cloud segmentation algorithm and region growing point cloud segmentation algorithm in terms of timeliness and correct segmentation rate. This new segmentation clustering algorithm can quickly implement automatic point cloud clustering segmentation of ground objects.

Point cloud segmentation of overhead contact systems with deep learning in high-speed rails

High-speed rails have strategic significance for political and economic development. As a critical part of the power supply system, overhead catenary systems (OCSs) power high-speed rails. OCSs are easily damaged due to artificial and natural factors. To ensure that OCSs work properly, researchers often use light detection and ranging (LiDAR) to recognize and inspect OCS components. However, recent OCS recognition methods relying on LiDAR need to enhance the ability to extract local features, integrate contextual features, and enhance crucial features while maintaining less latency and model complexity. To accomplish this objective, we proposed an accelerated point cloud segmentation algorithm for OCS recognition. The algorithm extracted local point cloud features based on machine learning methods. Then, it integrated the contextual features of point clouds using dot and accumulation operations. Additionally, we captured vital features and neglected unimportant features by constructing a feature enhancement algorithm. We accelerated the proposed algorithms by eliminating the run-time overhead of GPU scheduling. Experiments showed that our method had high precision with low model complexity and latency. For example, our precision was at least 0.64% better than comparison studies in recognizing steady arms. Our parameters were at least 43.59% fewer than others. Our optimized algorithm achieved a 2.47, 2.81, 3.24, 1.54, 7.12, and 7× speedup on the Nano, Tesla T4, RTX 2080 Ti, TX2, Tesla V100S GPU, and TITAN V, respectively. Our visualization effect also outperformed other methods in OCS recognition of high-speed rail scenarios.

Deep learning for large-scale point cloud segmentation in tunnels considering causal inference

This work designs an integrated deep learning approach to accomplish a multi-label segmentation of various objects including seepage from 3D point clouds of tunnels. 3D laser scanning is first utilized to collect raw point clouds from the operating tunnels. An integration approach based on projection strategy and point-based method is developed. A causal inference-based data augmentation approach is proposed to enhance the segmentation performance between the seepage and the segment class. A Dempster–Shafer (D-S) evidence-based feature fusion model is established to combine the local feature with global features from different CNN models. A real cross-river tunnel section in China is applied as a case study for illustration. Results verify that (1) the integration framework shows an outperforming performance of seepage segmentation, and the Intersection over Union (IOU), Precision, Recall, and F1 score for seepage segmentation achieve 0.70, 0.896, 0.762, and 0.824, respectively; (2) the segmentation model achieves competitive results in multiple classes, and the segmentation F1 score of the cable, segment, pipe, powertrack and the track classes obtain 0.913, 0.911, 0.993, 0.898, and 0.963, respectively; (3) the segmentation model is superior to other SOTA point cloud segmentation algorithms in terms of evaluation metrics.

Three-Dimensional Lidar Localization and Mapping with Loop-Closure Detection Based on Dense Depth Information

This paper presents a novel lidar SLAM system for localizing a mobile robot to build a map of the environment. To identify the unknown transform matrix, we design a new scan-matching approach, in which a point cloud segmentation algorithm is additionally integrated. Different from the traditional normal distribution transform algorithm for point cloud registration, our newly proposed one additionally incorporates a ground point remover and a point cloud segmentation method. By employing the point cloud segmentation algorithm to divide the point cloud space into different cells, the newly proposed algorithm can guarantee the continuity and convergence of the cost function. To tackle the recognition difficulties that the camera-based loop-closure detection heavily depends on the environment’s appearance, a depth-completion algorithm is introduced to fuse sensor data to ensure the robustness of the algorithm. Moreover, the bags of binary words (DBoW) are adopted to improve the image-matching quality. Finally, experimental results are presented to illustrate the effectiveness of the proposed system.

A Palletizing System for Microchannel Heat Exchangers Based on 3D Visual Guidance

Aiming at the problems of slow manual palletizing, high risk and easy damage to products after the assembly of microchannel heat exchanger, a palletizing system based on 3D visual guidance is presented in this paper. Firstly, an automatic calibration method for RGB-D camera extrinsic parameters was adopted: used random sample consensus algorithm RANSAC for plane fitting, and the pallet was idealized as a plane. The camera extrinsic parameters were calculated according to the rotation relationship between the plane normal vector and the camera coordinate system. When it was performing detection, converted depth data into point cloud data and preprocessed, including coordinate transformation, downsampling, etc. After that, according to the characteristics of stacking, a point cloud segmentation algorithm based on depth and the number of points was used, which could not only segment the pallet and sponge strips, but also effectively detect obstacles. For different parts segmented, the pallet was positioned by fitting the minimum bounding rectangle, and the stacking depths were obtained by calculating the centroid of the sponge strips. Finally, a series of constraints were used to determine whether the unloading conditions were met. The experimental results show that the automatic calibration is effective, the time consuming is about 0.43s, and the depth range of the pallet is 11mm after calibration. The algorithm can accurately identify various situations above the stacking and the time consuming is about 1.63s.

Point cloud segmentation algorithm fusing few-shot meta-learning and prototype alignment

Point cloud segmentation algorithm fusing few-shot meta-learning and prototype alignment

A point cloud segmentation and material statistics algorithm for train carriage

The efficiency of train transportation in the port environment directly restricts the production efficiency of the entire port. In order to coordinate the productivity of the entire port and further improve the loading and unloading efficiency of the train automation system, this paper proposes a material segmentation and data statistics algorithm for the problem of material identification in the train compartment. The terrestrial laser scanning system was established under the experimental scenario of CHN ENERGY Tianjin Port. By fusing the laser point cloud data, and then performing pre-processing operations such as box filtering, down-sampling, and radius filtering, the pre-processed point cloud data is projected onto a two-dimensional plane as a two-dimensional image, and the canny operator is used to extract the contour of the carriage on the image. Further fitting to the plane of the point cloud of the carriage wall, and the point cloud of materials in the carriage is segmented. Through the method of slicing the compartment, completing the statistical analysis of the data in the compartment from the whole to the part. The database fields of the main carriage types are established to realize the information interaction between the database and the recognition algorithm. The final experimental results show that the error rate of the carriage length recognition is 5.89%, and the width error rate is 7.25%, which verifies that the algorithm has good recognition accuracy which fully meets engineering needs.

An Advanced Framework for Multi-Scale Forest Structural Parameter Estimations Based on UAS-LiDAR and Sentinel-2 Satellite Imagery in Forest Plantations of Northern China

Regarded as a marked category of global forests, forest plantations not only have great significance for the development of the global economy, but also contribute ecological and social benefits. The accurate acquisition of the multi-scale (from individual tree to landscape level) and near-real-time information of structural parameters in plantations is the premise of decision-making in sustainable management for the whole forest farm, and it is also the basis for the evaluation of forest productivity in stands. The development and synergetic applications of multi-source and multi-platform remote sensing technology provide a technical basis for the highly accurate estimation of multi-scale forest structural parameters. In this study, we developed an advanced framework for estimating these parameters of forest plantations in multiple scales (individual tree, plot and landscape levels) based on the Unmanned Aircraft System Light Detection and Ranging (UAS-LiDAR) transects and wall-to-wall Sentinel-2 imagery, combined with the sample plot data in a typical forest farm plantation (mainly Larch, Chinese pine) of Northern China. The position and height of individual trees within the plots were extracted by the LiDAR-based point cloud segmentation (PCS) algorithm, and then different approaches to the extrapolation of forest structural parameters from the plot to landscape level were assessed. The results demonstrate that, firstly, the individual tree height obtained by PCS was of relatively high accuracy (rRMSE = 1.5–3.3%); secondly, the accuracy of the forest structure parameters of the sample plot scale estimated by UAS-LiDAR is rRMSE = 4.4–10.6%; and thirdly, the accuracy of the two-stage upscaling approach by UAS-LiDAR transects as an intermediate stage (rRMSE = 14.5–20.2%) performed better than the direct usage of Sentinel-2 data (rRMSE = 22.9–27.3%). This study demonstrated an advanced framework for creating datasets of multi-scale forest structural parameters in a forest plantation, and proved that the synergetic usage of UAS-LiDAR transects and full coverage medium-resolution satellite imagery can provide a high-precision and low-cost technical basis for the multi-level estimation of forest structural parameters.