Point Cloud Processing Methods for 3D Point Cloud Detection Tasks

Light detection and ranging(LiDAR)sensors play a vital role in acquiring 3D point cloud data and extracting valuable information about objects for tasks such as autonomous driving,robotics,and virtual reality(VR).However,the sparse and disordered nature of the 3D point cloud poses significant challenges to feature extraction.Overcoming limitations is critical for 3D point cloud processing.3D point cloud object detection is a very challenging and crucial task,in which point cloud processing and feature extraction methods play a crucial role and have a significant impact on subsequent object detection performance.In this overview of outstanding work in object detection from the 3D point cloud,we specifically focus on summarizing methods employed in 3D point cloud processing.We introduce the way point clouds are processed in classical 3D object detection algorithms,and their improvements to solve the problems existing in point cloud processing.Different voxelization methods and point cloud sampling strategies will influence the extracted features,thereby impacting the final detection performance.

Aggregate Point Cloud Geometric Features for Processing

As 3D acquisition technology develops and 3D sensors become increasingly affordable,large quantities of 3D point cloud data are emerging.How to effectively learn and extract the geometric features from these point clouds has become an urgent problem to be solved.The point cloud geometric information is hidden in disordered,unstructured points,making point cloud analysis a very challenging problem.To address this problem,we propose a novel network framework,called Tree Graph Network(TGNet),which can sample,group,and aggregate local geometric features.Specifically,we construct a Tree Graph by explicit rules,which consists of curves extending in all directions in point cloud feature space,and then aggregate the features of the graph through a cross-attention mechanism.In this way,we incorporate more point cloud geometric structure information into the representation of local geometric features,which makes our network perform better.Our model performs well on several basic point clouds processing tasks such as classification,segmentation,and normal estimation,demonstrating the effectiveness and superiority of our network.Furthermore,we provide ablation experiments and visualizations to better understand our network.

Building Facade Point Clouds Segmentation Based on Optimal Dual-Scale Feature Descriptors

To address the current issues of inaccurate segmentation and the limited applicability of segmentation methods for building facades in point clouds, we propose a facade segmentation algorithm based on optimal dual-scale feature descriptors. First, we select the optimal dual-scale descriptors from a range of feature descriptors. Next, we segment the facade according to the threshold value of the chosen optimal dual-scale descriptors. Finally, we use RANSAC (Random Sample Consensus) to fit the segmented surface and optimize the fitting result. Experimental results show that, compared to commonly used facade segmentation algorithms, the proposed method yields more accurate segmentation results, providing a robust data foundation for subsequent 3D model reconstruction of buildings.

Processing of Aerosol Particles in Convective Cumulus Clouds:Cases Study in the Mexican East Pacific

In the Mexican Intertropical Convergence Zone, particle size distributions within 500 m of cloud boundaries at altitudes of 1000, 2500, and 4200 m, were compared against size distributions at the same levels but 1500 m away from the clouds. The differences in the distributions near and far from the cloud are related to processes that may change particle properties inside the cloud. Chemical changes in the aerosols are deduced from the particles＇ refractive index, as derived from comparisons with the scattering coeflcient measured by a nephelometer. An analysis of ten cloud systems indicates that vertical transport of cloud base aerosol followed by entrainment/detrainment is the cloud processing signature most frequently observed in the comparisons （65%）. Changes in the chemical composition are observed in approximately 20% of the cases and another 20% of the cases showed removal by precipitation. About 5% of the comparisons showed clear evidence of changes by coalescence. The principal effect of these cloud-processed aerosols is observed in the increase of optical depth in the layer from 30 m to 4200 m in the near-cloud regions, in comparison with the atmosphere further from clouds.

Dominant Cloud Microphysical Processes of a Torrential Rainfall Event in Sichuan, China

High-resolution numerical simulation data of a rainstorm triggering debris flow in Sichuan Province of China simulated by the Weather Research and Forecasting （WRF） Model were used to study the dominant cloud microphysical processes of the torrential rainfall.The results showed that：（1） In the strong precipitation period,particle sizes of all hydrometeors increased,and mean-mass diameters of graupel increased the most significantly,as compared with those in the weak precipitation period; （2） The terminal velocity of raindrops was the strongest among all hydrometeors,followed by graupel＇s,which was much smaller than that of raindrops.Differences between various hydrometeors＇ terminal velocities in the strong precipitation period were larger than those in the weak precipitation period,which favored relative motion,collection interaction and transformation between the particles.Absolute terminal velocity values of raindrops and graupel were significantly greater than those of air upward velocity,and the stronger the precipitation was,the greater the differences between them were; （3） The orders of magnitudes of the various hydrometeors＇ sources and sinks in the strong precipitation period were larger than those in the weak precipitation period,causing a difference in the intensity of precipitation.Water vapor,cloud water,raindrops,graupel and their exchange processes played a major role in the production of the torrential rainfall,and there were two main processes via which raindrops were generated：abundant water vapor condensed into cloud water and,on the one hand,accretion of cloud water by rain water formed rain water,while on the other hand,accretion of cloud water by graupel formed graupel,and then the melting of graupel formed rain water.

Formal Modeling and Discovery of Multi-instance Business Processes: A Cloud Resource Management Case Study

Process discovery, as one of the most challenging process analysis techniques, aims to uncover business process models from event logs. Many process discovery approaches were invented in the past twenty years;however, most of them have difficulties in handling multi-instance sub-processes. To address this challenge, we first introduce a multi-instance business process model(MBPM) to support the modeling of processes with multiple sub-process instantiations. Formal semantics of MBPMs are precisely defined by using multi-instance Petri nets(MPNs)that are an extension of Petri nets with distinguishable tokens.Then, a novel process discovery technique is developed to support the discovery of MBPMs from event logs with sub-process multi-instantiation information. In addition, we propose to measure the quality of the discovered MBPMs against the input event logs by transforming an MBPM to a classical Petri net such that existing quality metrics, e.g., fitness and precision, can be used.The proposed discovery approach is properly implemented as plugins in the Pro M toolkit. Based on a cloud resource management case study, we compare our approach with the state-of-theart process discovery techniques. The results demonstrate that our approach outperforms existing approaches to discover process models with multi-instance sub-processes.

Research on 3D Laser Scanning Reconstruction of Ancient Buildings Combined with BIM Technology

After more than 30 years of scientific and social development, surveying and mapping technology by leaps and bounds, engineering surveying technology has undergone tremendous changes. In the process of protecting ancient buildings, it is necessary to obtain the precise dimensions of architectural details. In this study, the path of 3D laser scanning combined with BIM technology is explored. Taking the observation and protection of the ancestral hall of the Liu family as an example, this study aims to draw drawings that reflect the relevant information about the ancient buildings, the accurate three-dimensional model of ancient buildings is established with BIM technology, which provides new methods and ideas for the research and protection of ancient buildings. .

Dynamic A^*path finding algorithm and 3D lidar based obstacle avoidance strategy for autonomous vehicles

This paper presents a novel dynamic A^*path finding algorithm and 3D lidar based local obstacle avoidance strategy for an autonomous vehicle.3D point cloud data is collected and analyzed in real time.Local obstacles are detected online and a 2D local obstacle grid map is constructed at 10 Hz/s.The A^*path finding algorithm is employed to generate a local path in this local obstacle grid map by considering both the target position and obstacles.The vehicle avoids obstacles under the guidance of the generated local path.Experiment results have shown the effectiveness of the obstacle avoidance navigation algorithm proposed.

Assessment of handheld mobile terrestrial laser scanning for estimating tree parameters

Sustainable forest management heavily relies on the accurate estimation of tree parameters.Among others,the diameter at breast height(DBH) is important for extracting the volume and mass of an individual tree.For systematically estimating the volume of entire plots,airborne laser scanning(ALS) data are used.The estimation model is frequently calibrated using manual DBH measurements or static terrestrial laser scans(STLS) of sample plots.Although reliable,this method is time-consuming,which greatly hampers its use.Here,a handheld mobile terrestrial laser scanning(HMTLS) was demonstrated to be a useful alternative technique to precisely and efficiently calculate DBH.Different data acquisition techniques were applied at a sample plot,then the resulting parameters were comparatively analysed.The calculated DBH values were comparable to the manual measurements for HMTLS,STLS,and ALS data sets.Given the comparability of the extracted parameters,with a reduced point density of HTMLS compared to STLS data,and the reasonable increase of performance,with a reduction of acquisition time with a factor of5 compared to conventional STLS techniques and a factor of3 compared to manual measurements,HMTLS is considered a useful alternative technique.

Comparison of the Bright Band Characteristics Measured by Micro Rain Radar (MRR) at a Mountain and a Coastal Site in South Korea

Data from a long term measurement of Micro Rain Radar （MRR） at a mountain site （Daegwallyeong, DG, one year period of 2005） and a coastal site （Haenam, HN, three years 2004-2006） in South Korea were analyzed to compare the MRR measured bright band characteristics of stratiform precipitation at the two sites. On average, the bright band was somewhat thicker and the sharpness （average gradient of reflectivity above and below the reflectivity peak） was slightly weaker at DG, compared to those values at HN. The peak reflectivity itself was twice as strong and the relative location of the peak reflectivity within the bright band was higher at HN than at DG. Importantly, the variability of these values was much larger at HN than at DG. The key parameter to cause these differences is suggested to be the difference of the snow particle densities at the two sites, which is related to the degree of riming. Therefore, it is speculated that the cloud microphysical processes at HN may have varied significantly from un-rimed snow growth, producing low density snow particles, to the riming of higher density particles, while snow particle growth at DG was more consistently affected by the riming process, and therefore high density snow particles. Forced uplifting of cloudy air over the mountain area around DG might have resulted in an orographic supercooling effect that led to the enhanced riming of supercooled cloud drops.

A Gradient-Domain Based Geometry Processing Framework for Point Clouds

The use of point clouds is becoming increasingly popular. We present a general framework for performing geometry filtering on point-based surface through applying the meshless local Petrol-Galelkin （MLPG） to obtain the solution of a screened Poisson equation. The enhancement or smoothing of surfaces is controlled by a gradient scale parameter. Anisotropic filtering is supported by the adapted Riemannian metric. Contrary to the other approaches of partial differential equation for point-based surface, the proposed approach neither needs to construct local or global triangular meshes, nor needs global parameterization. It is only based on the local tangent space and local interpolated surfaces. Experiments demonstrate the efficiency of our approach.

A Modified Double-Moment Bulk Microphysics Scheme Geared toward the East Asian Monsoon Region

Representation of cloud microphysical processes is one of the key aspects of numerical models.An improved double-moment bulk cloud microphysics scheme(named IMY)was created based on the standard Milbrandt-Yau(MY)scheme in the Weather Research and Forecasting(WRF)model for the East Asian monsoon region(EAMR).In the IMY scheme,the shape parameters of raindrops,snow particles,and cloud droplet size distributions are variables instead of fixed constants.Specifically,the shape parameters of raindrop and snow size distributions are diagnosed from their respective shape-slope relationships.The shape parameter for the cloud droplet size distribution depends on the total cloud droplet number concentration.In addition,a series of minor improvements involving detailed cloud processes have also been incorporated.The improved scheme was coupled into the WRF model and tested on two heavy rainfall cases over the EAMR.The IMY scheme is shown to reproduce the overall spatial distribution of rainfall and its temporal evolution,evidenced by comparing the modeled results with surface gauge observations.The simulations also successfully capture the cloud features by using satellite and ground-based radar observations as a reference.The IMY has yielded simulation results on the case studies that were comparable,and in ways superior to MY,indicating that the improved scheme shows promise.Although the simulations demonstrated a positive performance evaluation for the IMY scheme,continued experiments are required to further validate the scheme with different weather events.

PCT:Point cloud transformer

The irregular domain and lack of ordering make it challenging to design deep neural networks for point cloud processing.This paper presents a novel framework named Point Cloud Transformer(PCT)for point cloud learning.PCT is based on Transformer,which achieves huge success in natural language processing and displays great potential in image processing.It is inherently permutation invariant for processing a sequence of points,making it well-suited for point cloud learning.To better capture local context within the point cloud,we enhance input embedding with the support of farthest point sampling and nearest neighbor search.Extensive experiments demonstrate that the PCT achieves the state-of-the-art performance on shape classification,part segmentation,semantic segmentation,and normal estimation tasks.

Regression Method for Rail Fastener Tightness Based on Center-Line Projection Distance Feature and Neural Network

In the railway system,fasteners have the functions of damping,maintaining the track distance,and adjusting the track level.Therefore,routine maintenance and inspection of fasteners are important to ensure the safe operation of track lines.Currently,assessment methods for fastener tightness include manual observation,acoustic wave detection,and image detection.There are limitations such as low accuracy and efficiency,easy interference and misjudgment,and a lack of accurate,stable,and fast detection methods.Aiming at the small deformation characteristics and large elastic change of fasteners from full loosening to full tightening,this study proposes high-precision surface-structured light technology for fastener detection and fastener deformation feature extraction based on the center-line projection distance and a fastener tightness regression method based on neural networks.First,the method uses a 3D camera to obtain a fastener point cloud and then segments the elastic rod area based on the iterative closest point algorithm registration.Principal component analysis is used to calculate the normal vector of the segmented elastic rod surface and extract the point on the centerline of the elastic rod.The point is projected onto the upper surface of the bolt to calculate the projection distance.Subsequently,the mapping relationship between the projection distance sequence and fastener tightness is established,and the influence of each parameter on the fastener tightness prediction is analyzed.Finally,by setting up a fastener detection scene in the track experimental base,collecting data,and completing the algorithm verification,the results showed that the deviation between the fastener tightness regression value obtained after the algorithm processing and the actual measured value RMSE was 0.2196 mm,which significantly improved the effect compared with other tightness detection methods,and realized an effective fastener tightness regression.

Comparison of heuristic and deep learning-based methods for ground classification from aerial point clouds

The automatic definition of the ground from 3D point clouds has been a common process for the last two decades,with many different approaches and applications that can be found in a vast literature.This paper presents a comparison of three different methodological concepts for ground classification,in order to establish the advantages and drawbacks of each method.First,a heuristic method,based on previous knowledge of the geometry and context of the 3D data.Secondly,a Deep Convolutional Network based on SegNet that classifies 2D images generated from the 3D point cloud.Finally,the third method applies a Deep Learning classification based on PointNet,which takes 3D points directly as inputs.To validate each method and compare them,public and labelled point clouds from the Actueel Hoogtebestand Nederland dataset are employed.Furthermore,the three methods are validated against the ISPRS 3D Semantic Labeling Contest benchmark.The results obtained show that the deep learning-based approaches outperform the heuristic method,with F-scores above 96%.The best results were obtained using a shallower version of SegNet,with F-score above 97%.

Impact of Cloud Microphysical Processes on the Simulation of Typhoon Rananim near Shore. Part I： Cloud Structure and Precipitation Features

By using the Advanced Regional Eta-coordinate Model （AREM）, the basic structure and cloud features of Typhoon Rananim are simulated and verified against observations. Five sets of experiments are designed to investigate the effects of the cloud microphysical processes on the model cloud structure and precipitation features. The importance of the ice-phase microphysics, the cooling effect related to microphysical characteristics change, and the influence of terminal velocity of graupel are examined. The results indicate that the cloud microphysical processes impact more on the cloud development and precipitation features of the typhoon than on its intensity and track. Big differences in the distribution pattern and content of hydro-meteors, and types and amount of rainfall occur in the five experiments, resulting in different heating and cooling effects. The largest difference of 24-h rain rate reaches 52.5 mm h-1 . The results are summarized as follows： 1） when the cooling effect due to the evaporation of rain water is excluded, updrafts in the typhoon＇s inner core are the strongest with the maximum vertical velocity of -19 Pa s-1 and rain water and graupel grow most dominantly with their mixing ratios increased by 1.8 and 2.5 g kg-1, respectively, compared with the control experiment; 2） the melting of snow and graupel affects the growth of rain water mainly in the spiral rainbands, but much less significantly in the eyewall area; 3） the warm cloud microphysical process produces the smallest rainfall area and the largest percentage of convective precipitation （63.19%）, while the largest rainfall area and the smallest percentage of convective precipitation （48.85%） are generated when the terminal velocity of graupel is weakened by half.

Impact of Cloud Microphysical Processes on the Simulation of Typhoon Rananim near Shore. Part II： Typhoon Intensity and Track

The impact of cloud microphysical processes on the simulated intensity and track of Typhoon Rananim is discussed and analyzed in the second part of this study. The results indicate that when the cooling effect due to evaporation of rain water is excluded, the simulated 36-h maximum surface wind speed of Typhoon Rananim is about 7 m s-1 greater than that from all other experiments; however, the typhoon landfall location has the biggest bias of about 150 km against the control experiment. The simulated strong outer rainbands and the vertical shear of the environmental flow are unfavorable for the deepening and maintenance of the typhoon and result in its intensity loss near the landfall. It is the cloud microphysical processes that strengthen and create the outer spiral rainbands, which then increase the local convergence away from the typhoon center and prevent more moisture and energy transport to the inner core of the typhoon. The developed outer rainbands are supposed to bring dry and cold air mass from the middle troposphere to the planetary boundary layer （PBL）. The other branch of the cold airflow comes from the evaporation of rain water itself in the PBL while the droplets are falling. Thus, the cut-off of the warm and moist air to the inner core and the invasion of cold and dry air to the eyewall region are expected to bring about the intensity reduction of the modeled typhoon. Therefore, the deepening and maintenance of Typhoon Rananim during its landing are better simulated through the reduction of these two kinds of model errors.

A Simulation Study on the Characteristics of Cloud Microphysics of Heavy Rainfall in the Meiyu Front

A heavy rainfall in the Meiyu front during 4-5 July 2003 is simulated by use of the non-hydrostatic mesoscale model MM5 （V3-6） with different explicit cloud microphysical parameterization schemes. The characteristics of microphysical process of convective cloud are studied by the model outputs. The simulation study reveals that： （1） The mesoscale model MM5 with explicit cloud microphysical process is capable of simulating the instant heavy rainfall in the Meiyu front, the rainfall simulation could be improved signifi- cantly as the model resolution is increased, and the Goddard scheme is better than the Reisner or Schultz scheme. （2） The convective cloud in the Meiyu front has a comprehensive structure composed of solid, liquid and vapor phases of water, the mass density of water vapor is the largest one in the cloud; the next one is graupel, while those of ice, snow, rain water and the cloud water are almost same. The height at which mass density peaks for different hydrometeors is almost unchangeable during the heavy rainfall period. The mass density variation of rain water, ice, and graupel are consistent with that of ground precipitation, while that of water vapor in the low levels is 1-2 h earlier than the precipitation. （3） The main contribution to the water vapor budget in the atmosphere is the convergence of vapor flux through advection and convection, which provides the main vapor source of the rainfall. Besides the basic process of the auto-conversion of cloud water to rain water, there is an additional cloud microphysical process that is essential to the formation of instant heavy rainfall, the ice-phase crystals are transformed into graupels first and then the increased graupels mix with cloud water and accelerates the conversion of cloud water to rain water. The positive feedback mechanism between latent heat release and convection is the main cause to maintain and develop the heavy precipitation.

THE MICROPHYSICAL STRUCTURE OF SNOW CLOUDS AND THE GROWTH PROCESS OF SNOW PARTICLES IN WINTER IN XINJIANG

The microphysical structure of snow clouds and the growth process of snow crystals were observed by means of instrumented aircraft, weather radar, snow crystal observations etc. in Urumqi region during the winter of 1982. The analysis of three cases show that about 70% of snow mass growth is produced in the lower layer below 2000 m under the cold front, and that the concentration of ice crystals is as high as 60 L^(-1) and the supercooled water is absent in lower clouds. We may infer that the deposition of ice crystals and the aggregation of snow crystals are important processes for the snow development. The microphysical structure of the snow band near the front aloft and its characteristics as a seeder cloud are also described in this paper.

Two-sided ultrasonic surface rolling process of aeroengine blades based on on-machine noncontact measurement

As crucial parts of an aeroengine,blades are vulnerable to damage from long-term operation in harsh environments.The ultrasonic surface rolling process(USRP)is a novel surface treatment technique that can highly improve the mechanical behavior of blades.During secondary machining,the nominal blade model cannot be used for secondary machining path generation due to the deviation between the actual and nominal blades.The clamping error of the blade also affects the precision of secondary machining.This study presents a two-sided USRP(TS-USRP)machining for aeroengine blades on the basis of on-machine noncontact measurement.First,a TS-USRP machining system for blade is developed.Second,a 3D scanning system is used to obtain the point cloud of the blade,and a series of point cloud processing steps is performed.A local point cloud automatic extraction algorithm is introduced to extract the point cloud of the strengthened region of the blade.Then,the tool path is designed on the basis of the extracted point cloud.Finally,an experiment is conducted on an actual blade,with results showing that the proposed method is effective and efficient.