3D Point Data Research Articles

Application of Remote Sensing Technologies for Assessing Planted Forests Damaged by Insect Pests and Fungal Pathogens: a Review

In this review, we highlight recent developments and applications in remote sensing that can improve the accuracy and timeliness of health assessments in plantations managed for timber and pulp production. The detection and mapping of damage extent and severity caused by insect pests and fungal pathogens is a common requirement of foresters managing plantations. The objectives of these surveys can range from early detection for targeted intervention to more strategic aims of predicting stand susceptibility or evaluating the performance of management strategies. Recent developments in remote sensing technologies and big data modelling techniques can now provide spatially explicit, quantitative solutions for these management objectives that are more accurate than manual field-based assessments of tree damage or airborne visual mapping. Past studies have identified a large number of spectral, textural and structural metrics that have been used in models to classify specific tree crown damage symptoms. This process requires a detailed understanding of the chronology of crown symptoms for specific damaging agents and the spectral responses to these symptoms. Continuing increases in the spatial and spectral resolution of remote sensors enables crown-level damage classification. The development of data processing workflows that fuse spectral information with three-dimensional (3D) data acquired simultaneously from single or different remote platforms promote the opportunities to derive both structural and physiological crown-level attributes that relate to crown damage. The simultaneous acquisition of spectral and 3D point data will enable plantation foresters to derive several spatial products, including the assessment of tree health in a cost-effective manner.

Shape-aware surface reconstruction from sparse 3D point-clouds.

The reconstruction of an object's shape or surface from a set of 3D points plays an important role in medical image analysis, e.g. in anatomy reconstruction from tomographic measurements or in the process of aligning intra-operative navigation and preoperative planning data. In such scenarios, one usually has to deal with sparse data, which significantly aggravates the problem of reconstruction. However, medical applications often provide contextual information about the 3D point data that allow to incorporate prior knowledge about the shape that is to be reconstructed. To this end, we propose the use of a statistical shape model (SSM) as a prior for surface reconstruction. The SSM is represented by a point distribution model (PDM), which is associated with a surface mesh. Using the shape distribution that is modelled by the PDM, we formulate the problem of surface reconstruction from a probabilistic perspective based on a Gaussian Mixture Model (GMM). In order to do so, the given points are interpreted as samples of the GMM. By using mixture components with anisotropic covariances that are "oriented" according to the surface normals at the PDM points, a surface-based fitting is accomplished. Estimating the parameters of the GMM in a maximum a posteriori manner yields the reconstruction of the surface from the given data points. We compare our method to the extensively used Iterative Closest Points method on several different anatomical datasets/SSMs (brain, femur, tibia, hip, liver) and demonstrate superior accuracy and robustness on sparse data.

RGB-D Camera Based Walking Pattern Recognition by Support Vector Machines for a Smart Rollator.

This paper presents a walking pattern detection method for a smart rollator. The method detects the rollator user's lower extremities from the depth data of an RGB-D camera. It then segments the 3D point data of the lower extremities into the leg and foot data points, from which a skeletal system with 6 skeletal points and 4 rods is extracted and used to represent a walking gait. A gait feature, comprising the parameters of the gait shape and gait motion, is then constructed to describe a walking state. K-means clustering is employed to cluster all gait features obtained from a number of walking videos into 6 key gait features. Using these key gait features, a walking video sequence is modeled as a Markov chain. The stationary distribution of the Markov chain represents the walking pattern. Three Support Vector Machines (SVMs) are trained for walking pattern detection. Each SVM detects one of the three walking patterns. Experimental results demonstrate that the proposed method has a better performance in detecting walking patterns than seven existing methods.

Affine registration of point clouds based on point-to-plane approach

The problem of aligning of 3D point data is the known registration task. The most popular registration algorithm is the Iterative Closest Point (ICP). This paper proposes a new algorithm for affine registration of point clouds by incorporating the affine transformation into the point-to-plane ICP algorithm. At each iterative step of the algorithm, a closed-form solution for the affine transformation is derived.

A Study on Airborne LiDAR Calibration and Operation Techniques for Bathymetric Survey

The necessity of maritime sector for continuous management, accurate and update location information such as seabed shape and location, research on airborne LiDAR bathymetry surveying techniques are accelerating. Airborne LiDAR systems consist of a scanner and GPS/INS. The location accuracy of 3D point data obtained by a LiDAR system is determined by external orientation parameters. However, there are problems in the synchronization between sensors should be performed due to a variety of sensor combinations and arrangement. To solve this issue, system calibration should be conducted. Therefore, this study evaluates the system verification methods, processes, and operation techniques.

AIRBORNE MULTISPECTRAL LIDAR DATA FOR LAND-COVER CLASSIFICATION AND LAND/WATER MAPPING USING DIFFERENT SPECTRAL INDEXES

Airborne Light Detection And Ranging (LiDAR) data is widely used in remote sensing applications, such as topographic and landwater mapping. Recently, airborne multispectral LiDAR sensors, which acquire data at different wavelengths, are available, thus allows recording a diversity of intensity values from different land features. In this study, three normalized difference feature indexes (NDFI), for vegetation, water, and built-up area mapping, were evaluated. The NDFIs namely, NDFI<sub>G-NIR</sub>, NDFI<sub>G-MIR</sub>, and NDFI<sub>NIR-MIR</sub> were calculated using data collected at three wavelengths; green: 532 nm, near-infrared (NIR): 1064 nm, and mid-infrared (MIR): 1550 nm by the world’s first airborne multispectral LiDAR sensor “Optech Titan”. The Jenks natural breaks optimization method was used to determine the threshold values for each NDFI, in order to cluster the 3D point data into two classes (water and land or vegetation and built-up area). Two sites at Scarborough, Ontario, Canada were tested to evaluate the performance of the NDFIs for land-water, vegetation, and built-up area mapping. The use of the three NDFIs succeeded to discriminate vegetation from built-up areas with an overall accuracy of 92.51%. Based on the classification results, it is suggested to use NDFI<sub>G-MIR</sub> and NDFI<sub>NIR-MIR</sub> for vegetation and built-up areas extraction, respectively. The clustering results show that the direct use of NDFIs for land-water mapping has low performance. Therefore, the clustered classes, based on the NDFIs, are constrained by the recorded number of returns from different wavelengths, thus the overall accuracy is improved to 96.98%.

AIRBORNE MULTISPECTRAL LIDAR DATA FOR LAND-COVER CLASSIFICATION AND LAND/WATER MAPPING USING DIFFERENT SPECTRAL INDEXES

Abstract. Airborne Light Detection And Ranging (LiDAR) data is widely used in remote sensing applications, such as topographic and landwater mapping. Recently, airborne multispectral LiDAR sensors, which acquire data at different wavelengths, are available, thus allows recording a diversity of intensity values from different land features. In this study, three normalized difference feature indexes (NDFI), for vegetation, water, and built-up area mapping, were evaluated. The NDFIs namely, NDFIG-NIR, NDFIG-MIR, and NDFINIR-MIR were calculated using data collected at three wavelengths; green: 532 nm, near-infrared (NIR): 1064 nm, and mid-infrared (MIR): 1550 nm by the world’s first airborne multispectral LiDAR sensor “Optech Titan”. The Jenks natural breaks optimization method was used to determine the threshold values for each NDFI, in order to cluster the 3D point data into two classes (water and land or vegetation and built-up area). Two sites at Scarborough, Ontario, Canada were tested to evaluate the performance of the NDFIs for land-water, vegetation, and built-up area mapping. The use of the three NDFIs succeeded to discriminate vegetation from built-up areas with an overall accuracy of 92.51%. Based on the classification results, it is suggested to use NDFIG-MIR and NDFINIR-MIR for vegetation and built-up areas extraction, respectively. The clustering results show that the direct use of NDFIs for land-water mapping has low performance. Therefore, the clustered classes, based on the NDFIs, are constrained by the recorded number of returns from different wavelengths, thus the overall accuracy is improved to 96.98%.

Building Zone Regulation Compliance Using LIDAR Data: Real-Life Tests in İstanbul

Airborne Laser scanning systems with light detection and ranging (LIDAR) technology is one of the fast and accurate 3D point data acquisition techniques. Generating accurate digital terrain and/or surface models (DTM/DSM) is the main application of collecting LIDAR range data. The LIDAR technique facilitates the rapid production of models which contain ground and above surface objects in 3D (Jaafar et al., 1999). Today automatic / semi-automatic generation of 3D digital building models and other city furniture from LIDAR data is a very active area of scientific research (Elaksher and Bethel, 2002) indicates that manual surface reconstruction is very costly and time consuming, and the development of automated algorithms is of great importance. Currently with this LIDAR data in place, it is possible to derive models and information (e.g. the heights of the buildings).This paper focuses on and explains our efforts on automatically checking Buildings’ Zoning Regulation Compliance by integrating geometric information derived from 3D LIDAR data and semantic information acquired from 2D Implementation Development Maps

Application of Terrestrial 3D Laser Scanning to Monitor Changes of Beach Landforms

ABSTRACT Lee, H.S.; Kim, I.H., and Kim, H.G., 2016. Application of terrestrial 3D laser scanning to monitor changes of beach landforms. In: Vila-Concejo, A.; Bruce, E.; Kennedy, D.M., and McCarroll, R.J. (eds.), Proceedings of the 14th International Coastal Symposium (Sydney, Australia). Journal of Coastal Research, Special Issue, No. 75, pp. 173–177. Coconut Creek (Florida), ISSN 0749-0208. Monitoring changes in the beach morphology of coastal landforms is important when considering coastal management measures. In this paper, to create a changing beach surface, point cloud data of the beach are obtained using three-dimensional (3D) terrestrial laser scanning (TLS), and a beach surface model is analysed based on 3D point data. The 3D point cloud is generated from the scanned beach, including breakwaters, and these points are registered and merged through a reference point (scan origin and ball target) surveyed by RTK-GPS. Noise elements and unnecessary points are eliminated to yield better surface modelli...

레이저 구조광 영상기반 3차원 스캐너 개발

This paper addresses the development of 3D data acquisition system (3D scanner) based laser structured-light image. The 3D scanner consists of a stripe laser generator, a conventional camera, and a rotation table. The stripe laser onto an object has distortion according to 3D shape of an object. By analyzing the distortion of the laser stripe in a camera image, the scanner obtains a group of 3D point data of the object. A simple semiconductor stripe laser diode is adopted instead of an expensive LCD projector for complex structured-light pattern. The camera has an optical filter to remove illumination noise and improve the performance of the distance measurement. Experimental results show the 3D data acquisition performance of the scanner with less than 0.2㎜ measurement error in 2 minutes. It is possible to reconstruct a 3D shape of an object and to reproduce the object by a commercially available 3D printer.

Voxelization algorithms for geospatial applications: Computational methods for voxelating spatial datasets of 3D city models containing 3D surface, curve and point data models

Voxel representations have been used for years in scientific computation and medical imaging. The main focus of our research is to provide easy access to methods for making large-scale voxel models of built environment for environmental modelling studies while ensuring they are spatially correct, meaning they correctly represent topological and semantic relations among objects.In this article, we present algorithms that generate voxels (volumetric pixels) out of point cloud, curve, or surface objects. The algorithms for voxelization of surfaces and curves are a customization of the topological voxelization approach [1]; we additionally provide an extension of this method for voxelization of point clouds.The developed software has the following advantages:•It provides easy management of connectivity levels in the resulting voxels.•It is not dependant on any external library except for primitive types and constructs; therefore, it is easy to integrate them in any application.•One of the algorithms is implemented in C++ and C for platform independence and efficiency.

스테레오 영상을 이용한 비행 중인 항공기 날개의 변위 및 진동 측정

The present study was conducted by using stereo pattern recognition method(SPR method) to measure the displacement and vibration of an airplane wing in flight condition. A SPR based measurement system was developed using two visible light stereo cameras. The visible light stereo images were processed to obtain marker points by adaptive threshold method and marker filtering technique. The marker points were used to reconstruct 3D point, displacement, and vibration data. The SPR system was installed on F-16 fighter. The wing displacement and vibration were measured in flight condition. Therefore, this paper presents a possibility that SPR based measurement system using visible light stereo camera can be very useful for measuring displacement and vibration of an airplane in flight condition.

SU‐E‐J‐160: Motion Analysis and 3D Plane Fit Through the Tumor Center of Mass Positions in the 4DCT Data Set

Purpose:To model GTV's center of mass (COM) motion planes based on 4DCT lung data sets and establish possible correlation with the GTV's size and/or location as an aid in defining adequate GTV margins based on the most prominent motion amplitude in the 3D space.Methods:For the modeling of the COM motion throughout the respiratory phases, eleven lung cancer patients were used and the least square fitting was applied. The least square fitting generally involves either data smoothening or apparent trend identification. Quadratic minimization, singular value decomposition, orthogonal projections (orthogonal regression), are ways to find a least square solution. The linear least squares method was initially used for the planar fitting of the 3D point data. Evaluation of the goodness of fit was performed through the root mean square deviations (root mean square error, RMSE) and normalized RMSE (NRMSE) values.Results:The summed square of residuals of the fit method varied from 0.02 to 0.26 cm and the coefficient of determination from 0.04 to 0.89, while RMSE ranged from 0.05 to 0.13 and NRMSE from 0.11 to 0.29. The most prominent motion was in the transverse and sagittal planes. A correlation was found between the best fit motion plane and the GTV size and/or location. In the lower GTVs, when maximum SI was 1.5, AP and RL motions were 0.87 and 1.0 cm, respectively. This Result confirms the need for margins larger than 0.5 cm for these GTVs.Conclusion:The lower RMSE and NRMSE values showed that for each patient a good planar fit can be achieved for the COM motion path. Knowing the GTV size and location, a good estimate of the motion plane in the 3D space can be made. The required dosimetric margins can be easily calculated knowing only one motion extent and motion plane angle.

Geometric features for robust registration of point clouds

Several feature detectors for 3D point data have been proposed in the literature. They have been applied to various problems in computer vision and robotics. We use them to solve two fundamental problems in real-time robotics, namely the registration of laser scans as well as the detection of loops and places. We extend and modify existing feature detectors, combine them in a smart way and create a system, that solves these problems efficiently and better that existing other solutions. We evaluate our system with data sets provided by other groups as well as our own data and we compare our results to those obtained with other algorithms.

Nasolabial symmetry and esthetics in cleft lip and palate: analysis of 3D facial images.

To determine the relationship between nasolabial symmetry and esthetics in subjects with orofacial clefts. Eighty-four subjects (mean age 10 years, standard deviation 1.5) with various types of nonsyndromic clefts were included: 11 had unilateral cleft lip (UCL); 30 had unilateral cleft lip and alveolus (UCLA); and 43 had unilateral cleft lip, alveolus, and palate (UCLAP). A 3D stereophotogrammetric image of the face was taken for each subject. Symmetry and esthetics were evaluated on cropped 3D facial images. The degree of asymmetry of the nasolabial area was calculated based on all 3D data points using a surface registration algorithm. Esthetic ratings of various elements of nasal morphology were performed by eight lay raters on a 100 mm visual analog scale. Statistical analysis included ANOVA tests and regression models. Nasolabial asymmetry increased with growing severity of the cleft (p = 0.029). Overall, nasolabial appearance was affected by nasolabial asymmetry; subjects with more nasolabial asymmetry were judged as having a less esthetically pleasing nasolabial area (p < 0.001). However, the relationship between nasolabial symmetry and esthetics was relatively weak in subjects with UCLAP, in whom only vermilion border esthetics was associated with asymmetry. Nasolabial symmetry assessed with 3D facial imaging can be used as an objective measure of treatment outcome in subjects with less severe cleft deformity. In subjects with more severe cleft types, other factors may play a decisive role. Assessment of nasolabial symmetry is a useful measure of treatment success in less severe cleft types.

A cylindrical neighborhood for multi-view range images processing

The processing technologies of multi-view range images (MRIs) scanned by three-dimensional (3D) laser scanners, structured light scanners and laser radars (Ladars) are current research hotspot. Most of the processing such as denoising, integration and surface reconstruction are performed in local neighborhood. Thus, fast and precise query of the local neighborhoods of the 3D point data in MRIs is very important. Due to measurement accuracy and alignment error, there are gaps in overlapping regions and through theoretical analysis and experimental verification, we notice that the gaps which may result in loss of local neighborhood information have great influence on the neighborhood query whereas the previous methods ignore it. Thereby, a novel local neighborhood which is named cylindrical neighborhood (CYND) is proposed. Its searching procedures are described in detail. Compared with the previous methods, our CYND expands the searching scope along the directions of normal vectors of the 3D point data, hence eliminating the impact of gaps. The widely used and well-implemented ANN library was employed for a comparison study and the experimental results demonstrate the validity and superiority of our CYND. With extensive application of 3D optical scanning technologies, this novel neighborhood will have a wide application prospect.

Raster Vs. Point Cloud LiDAR Data Classification

Abstract. Airborne Laser Scanning systems with light detection and ranging (LiDAR) technology is one of the fast and accurate 3D point data acquisition techniques. Generating accurate digital terrain and/or surface models (DTM/DSM) is the main application of collecting LiDAR range data. Recently, LiDAR range and intensity data have been used for land cover classification applications. Data range and Intensity, (strength of the backscattered signals measured by the LiDAR systems), are affected by the flying height, the ground elevation, scanning angle and the physical characteristics of the objects surface. These effects may lead to uneven distribution of point cloud or some gaps that may affect the classification process. Researchers have investigated the conversion of LiDAR range point data to raster image for terrain modelling. Interpolation techniques have been used to achieve the best representation of surfaces, and to fill the gaps between the LiDAR footprints. Interpolation methods are also investigated to generate LiDAR range and intensity image data for land cover classification applications. In this paper, different approach has been followed to classifying the LiDAR data (range and intensity) for land cover mapping. The methodology relies on the classification of the point cloud data based on their range and intensity and then converted the classified points into raster image. The gaps in the data are filled based on the classes of the nearest neighbour. Land cover maps are produced using two approaches using: (a) the conventional raster image data based on point interpolation; and (b) the proposed point data classification. A study area covering an urban district in Burnaby, British Colombia, Canada, is selected to compare the results of the two approaches. Five different land cover classes can be distinguished in that area: buildings, roads and parking areas, trees, low vegetation (grass), and bare soil. The results show that an improvement of around 10 % in the classification results can be achieved by using the proposed approach.

Rapid development of complex shaped customized products

In this work, two complex-shaped products have been reverse engineered through rapid prototyping (RP). First product is a single unit with a number of complex features while the second product has sub-parts to be assembled together. These sub-parts have various complexities. One sub-part will frequently be in motion. Target points are set on complex features of the scanned 3D point data to measure deviations of generated surface models from scanned data and subsequently from physical models. Two layer-by-layer manufacturing systems, i.e. one employing ceramic powder-based 3D printing and other fused deposition method have been used to develop RP models of complex shapes. Suitability of the process for complex-shaped prototypes has been discussed and recommendations have been made for parameter settings and suitable process. Moreover, economic analysis has also been performed with an objective to minimize the development time and cost.

The Algorithm of Obtaining 3D Point Cloud Data in Ground Fissure

In this paper, we focus on the problem of how to obtain 3D cloud point data from different data source in ground fissure. We mainly analyzed the spatial morphology and the data characteristics of the ground fissure. The curve formula is constructed in engineering geological section and designed the algorithm of obtaining 3D point cloud data. This algorithm has been verified in Xi’an area and obtained satisfactory results. The research achievement provides a new method for using multi-source data on 3D modeling ground fissure.

Parametric generation of planing hulls

This paper presents a mathematical method for producing hard-chine ship hulls based on a set of numerical parameters that are directly related to the geometric features of the hull and uniquely define a hull form for this type of ship. The term planing hull is used generically to describe the majority of hard-chine boats being built today. This paper is focused on un-stepped, single-chine hulls. B-spline curves and surfaces were combined with constraints on the significant ship curves to produce the final hull design. The hard-chine hull geometry was modeled by decomposing the surface geometry into boundary curves, which were defined by design constraints or parameters. In planing hull design, these control curves are the center, chine and sheer lines, as well as their geometric features including position, slope and, in the case of the chine, enclosed area and centroid. These geometric parameters have physical, hydrodynamic and stability implications from the design point of view. The proposed method utilizes 2D orthogonal projections of the control curves and then produces 3D definitions using B-spline fitting of the 3D data points. The fitting considers maximum deviation from the curve to the data points and is based on an original selection of the parameterization. A net of B-spline curves (stations) is then created to match the previously defined 3D boundaries. A final set of lofting surfaces of the previous B-spline curves produces the hull surface.