Search

The analysis of topographic time series is often based on bitemporal change detection and quantification. For 3D point clouds, acquired using laser scanning or photogrammetry, random and systematic noise has to be separated from the signal of surface change by determining the minimum detectable change. To analyse geomorphic change in point cloud data, the multiscale model-to-model cloud comparison (M3C2) approach is commonly applied, which provides a statistical significance test. This test assumes planar surfaces and a uniform registration error. For natural surfaces, the planarity assumption does not necessarily apply, in which cases the value of minimal detectable change (Level of Detection) is overestimated. To overcome these limitations, we quantify an uncertainty information for each 3D point by propagating the uncertainty of the measurements themselves and of the alignment uncertainty to the 3D points. This allows the calculation of 3D covariance information for the point cloud, which we use in an extended statistical test for equality of multivariate means. Our method, called M3C2-EP, gives a less biased estimate of the Level of Detection, allowing a more appropriate significance threshold in typical cases. We verify our method in two simulated scenarios, and apply it to a time series of terrestrial laser scans of a rock glacier at two different timespans of three weeks and one year. Over the three-week period, we detect significant change at 12.5% fewer 3D locations, while quantifying additional 25.2% of change volume, when compared to the reference method of M3C2. Compared with manual assessment, M3C2-EP achieves a specificity of 0.97, where M3C2 reaches 0.86 for the one-year timespan, while sensitivity drops from 0.72 for M3C2 to 0.60 for M3C2-EP. Lower Levels of Detection enable the analysis of high-frequency monitoring data, where usually less change has occurred between successive scans, and where change is small compared to local roughness. Our method further allows the combination of data from multiple scan positions or data sources with different levels of uncertainty. The combination using error propagation ensures that every dataset is used to its full potential.

Airborne laser scanning (ALS) has emerged as a technology capable of generating descriptors of vegetation structure and best available terrain information. Research and operational implementations of ALS data have highlighted their value for characterizing forest structure and generating spatially explicit and objective spatial coverages and mapping products for forest management. Continued emphasis to enhance forest stewardship is promoting novel methods to integrate ALS to detail non-timber ecosystem values like habitat, soil, and water. Standardized criteria and indicator frameworks such as the Canadian Council of Forest Ministers provide a reliable starting point for where ALS has opportunities to characterize ecosystems objectively regardless of location. In this review of primarily Canadian work, we highlight how ALS is becoming an increasingly viable technology for deriving meaningful indicators to meet sustainable forest management criteria. We review and highlight the value of ALS for quantifying indicators of biological diversity, ecosystem condition and productivity, soil and water, and the role of forests in global ecological cycles. We conclude by highlighting the need for increased education, tech transfer, flexible software, and reporting frameworks alongside five key considerations for using ALS to derive meaningful indicators of sustainable forest management.

Forest management inventories assisted by airborne laser scanning (ALS) can be used to predict different forest attributes. These predictions are utilized in practical forestry, but in the case of timber assortment-specific volumes, the ALS-based predictions can be inaccurate. This causes uncertainty in harvest planning. However, ALS-based predictions can be calibrated to achieve greater accuracy with local measurements. In this study, we used ALS data and accurately positioned cut-to-length harvester measurements from Norway spruce (Picea abies (L.) Karst.) dominated clear-cuts. We fitted linear mixed-effects (LME) models with exponential correlation structure for merchantable volume and sawlog volume for 225 m2 cells. Our aim was to study the effect of local harvester measurements on the accuracy of stand level merchantable and sawlog volumes. LME-based predictions were calibrated repeatedly up to 40 times as the cutting progressed. ALS data and harvester measurements were used to predict both the random effects and residual errors for each validation unit. At best, relative root mean square error (RMSE%) of initial predictions of 15.4 per cent for merchantable volume and 22.1 per cent for sawlog volume were reduced to 4.1 and 5.3 per cent, respectively, when measurements from 40 harvested cells of size 225 m2 were used. These results suggest that spatially accurate harvester data could be utilized during harvesting to increase the accuracy of volume and timber assortment predictions.

Various means of extracting road boundary from mobile laser scanning data based on vehicle trajectories have been investigated. Independent of positioning and navigation data, this study estimated the scanner ground track from the spatial distribution of the point cloud as an indicator of road location. We defined a typical edge block consisting of multiple continuous upward fluctuating points by abrupt changes in elevation, upward slope, and road horizontal slope. Subsequently, such edge blocks were searched for on both sides of the estimated track. A pseudo-mileage spacing map was constructed to reflect the variation in spacing between the track and edge blocks over distance, within which road boundary points were detected using a simple linear tracking model. Experimental results demonstrate that the ground trajectory of the extracted scanner forms a smooth and continuous string just on the road; this can serve as the basis for defining edge block and road boundary tracking algorithms. The defined edge block has been experimentally verified as highly accurate and strongly noise resistant, while the boundary tracking algorithm is simple, fast, and independent of the road boundary model used. The correct detection rate of the road boundary in two experimental data is more than 99.2%.

The purpose of this work is to develop a methodology for monitoring real estate objects, as a single information block, by including this procedure for capital construction objects in the process of land use control to update the information contained in state systems. The object of the research is the systems of state monitoring of land use based on the automated information system of state monitoring of land and registration of immovable objects based on the federal state information system of the Unified State Register of Real Estate. The subject of the work is creation a concept for a system of monitoring immovable objects, as a single resource, by introducing a monitoring unit for capital construction ones. As a result of the research, the stages of the method implementation are described, the key users of the final monitoring system for real estate objects are presented and options for its practical application are indicated.

The use of airborne laser scanning (LS) is increasing in forestry. Scanning can be conducted from manned aircrafts or unmanned aerial vehicles (UAV). The scanning data are often used to calculate various attributes for small raster cells. These attributes can be used to segment the forest into homogeneous areas, called segments, micro-stands, or, like in this study, stands. Delineation of stands from raster data is equal to finding the most suitable stand number for each raster cell, which is a combinatorial optimization problem. This study tested the performance of the simulated annealing (SA) metaheuristic in the delineation of stands from grids of UAV-LS attributes. The objective function included three criteria: within-stand variation of the LS attributes, stand area, and stand shape. The purpose was to create delineations that consisted of homogeneous stands with a low number of small stands and a regular and roundish stand shape. The results showed that SA is capable of producing stand delineations that meet these criteria. However, the method tended to produce delineations where the stands often consisted of disconnected parts and the stand borders were jagged. These problems were mitigated by using a mode filter on the grid of stand numbers and giving unique numbers for all disconnected parts of a stand. Three LS attributes were used in the delineation. These attributes described the canopy height, the height of the bottom of the canopy and the variation of echo intensity within 1-m2 raster cells. Besides, a texture variable that described the spatial variation of canopy height in the proximity of a 1-m2 raster cell was found to be a useful variable. Stand delineations where the average stand area was about one hectare explained more than 80% of the variation in canopy height.

The paper presents an application of the methodology used in the paper “Synthetic analysis of geoinformatics technologies for cultural heritage conservation, methodological approach”. The creation of the 3D model of the Rákóczi-Bánffy Castle in Urmeniș, Bistrița Năsăud County, was done by applying and integrating Terrestrial Laser Scanning (TLS) technologies and aerial photogrammetry performed with an Unmanned Aerial Vehicle (UAV). Agisoft Photoscan was used to compare the results and then they were compared with the images scanned using CloudCompare software. Thus, following the performance, with the help of the mentioned software, of a series of processing of the point cloud obtained, through the two imaging technologies, the error between the points belonging to the point cloud taken with UAV and the one taken with Laser Scanner was between 1 and 15 mm, the margin of error being acceptable for monuments without complex architectural details, so that the point cloud resulting from UAVs can be used successfully in this activity. The aim of the paper is to elaborate a geomatic methodology with an optimized cost-quality ratio, later replicable in the analysis of the current state of other constructions of the same type, knowing that over 600 castles in Transylvania alone are in a similar state, and such cases can be found in other Central European states as well.

The perfect knowledge about the scanned object’s geometry is essential for the empirical analysis of the stochastic properties of terrestrial laser scanners (TLSs). The Bonn reference wall is intended to be used as a reference for TLS quality investigations. Therefore, it is necessary to know the geometry of the wall at each time of scanning to avoid the misinterpretation of possible movements as systematic effects in the scanner. For this reason, we investigate the stability of the Bonn reference wall in this study. This includes the definition of a geodetic datum, the quantification of displacements, and the establishment of a suited deformation model. Since we discover a movement of about 1 mm within one day and up to 7 mm over the year, it is necessary to establish a cause-response deformation model to correct the wall movements in the scans. This study proposes two dynamic deformation models to compensate for the movements of the wall within one day and within a year. Our results show that it is better to measure the initial geometry of the wall each day since 89 % of the relative movements can be reduced to a maximum of 0.25 mm with a standard deviation of 0.16 mm (0.23 mm without modeling). If the shape is not initially known each day, the standard deviation of the displacements can be reduced from 1.10 mm to 0.61 mm, but the largest residuals still amount up to 2.5 mm, which is not sufficient for stochastic TLS investigations.

Forest inventory based on airborne laser scanning produces data for small grid cells. Various segmentation methods are used to cluster the grid cells into spatially continuous larger areas that cor...

The measurement of surface moisture on beaches is vital for studying aeolian sand transport mechanisms, but existing techniques are not adequate for monitoring the surface moisture dynamics over a substantial beach section. In this study, we investigated the suitability of a new remote sensing method to monitor the spatiotemporal variation in surface moisture on a sandy beach using a long-range static terrestrial laser scanner (TLS). The TLS was permanently deployed on top of a 42 m high building overlooking the study site at Ostend-Mariakerke, Belgium. Considering the effect of target surface roughness on the intensity and the laboratory’s length limitation, a new intensity correction method is proposed which only uses the field point cloud data measured on a homogenous beach surface (without time-consuming indoor experiments). Based on the corrected intensity data, the relation of the beach surface moisture to the corrected intensity was modeled by an exponential model with a correlation-coefficient squared of 0.92. A moisture estimation model was developed which can directly derive the beach surface moisture from the original intensity data of the TLS with a standard error of 2.27%. The hourly surface moisture dynamics across two tidal cycles on the beach were investigated as a case study, in which the point clouds derived from corresponding unmanned aerial vehicle (UAV) imagery are utilized to improve the calculation accuracy of the incidence angles of TLS point clouds at long distances. Results reveal that, after the intensity correction, the long-range static TLS is an extremely suitable technique to monitor the surface moisture dynamics (daytime and nighttime) over a substantial beach section (hundreds of meters) at a high scanning frequency (minutes to hours).

Alongside spatial information, an intensity scalar is also measured by LiDAR (Light Detection And Ranging) sensor systems. It corresponds to the amplitude of the backscattered laser beam after reflection on the scanned surface. This information isn’t directly usable due to dependencies of geometrical parameters occurring during the scanning process and additional processing modifications. The research community, in a growing trend, invests efforts to convert this signal into a value which could bring qualitative and quantitative new applications. We propose in this paper, a review of the theortical background of LiDAR radiometric calibration, from physical background to state-of-art methodology. A comparison of two approaches to eliminate geometrical (range and incidence angle) and instrumental dependencies of intensity measurement is presented. Their application on several homogeneous areas of a real-world case study shows a significant reduction of intensity variation between surfaces with identical material composition. Finally, a linearization process permits to obtain an equivalent Lambertian reflectance value which, by its disciminant potential, could be specifically suitable for some algorithms like segmentation or registration.

Tree improvement programs are critical to establishing high yield seed sources while maintaining genetic diversity and developing sustainable plantation forests. Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) is commonly used in improvement programs due to its superior strength and stiffness properties. Progeny testing trials in British Columbia (BC), Canada aim to increase stem volume without sacrificing wood quality. As genetically improved stock in BC is entering its fourth generation, wood quality and branch attributes are becoming more important as selection criteria evolve. This study investigates the utility of Airborne Laser Scanning (ALS) to produce metrics that describe branch attributes, and test whether these attributes showed differences in trees of three different genetic gain levels (GL); Wild-Stand (WS), Mid-Gain (MG) and Top-Cross (TC), planted at two different spacings (2.9 m and 4.0 m) in a realized-gain trial. New methods were developed to utilize ALS data to estimate metrics such as branch angle, length and volume using a point clustering approach. The relative impact of GL and spacing on branch attributes were assessed. Spacing was significant for branch angle (2.9 m = 73.53°, 4.0 m = 72.46°), whereas GL (WS = 0.861 m, MG = 0.917 m, TC = 0.948 m) and spacing (2.9 m = 0.884, 4.0 m = 0.942) were significant for length. For all metrics, TC trees at 4.0 m spacing were consistently significantly different whenever GL or the interaction effect was significant. This data provide an insight into how ALS can be used to model branch attributes, whereas the ability to analyse trees by plot, individual tree and individual branch attributes further allows researchers and foresters to maximize the value of ALS data. Findings from this research can be integrated into large-scale programs not just for monitoring trees, but also for identifying new trees that can display attributes associated with larger volumes and increased value.

Airborne laser scanning (ALS; LiDAR) data are an increasingly common data source for forest inventories, and approaches integrating ALS data with field plot measurements have become operational in several jurisdictions. As technology continues to evolve, different LiDAR sensors can provide new opportunities to incorporate LiDAR data into forest inventory workflows. Single photon LiDAR (SPL) enables efficient, large area data acquisition and merits further investigation for forest inventory applications. Herein, we investigated the capacity of leaf-on SPL data, combined with 269 field plots, for estimating forest inventory attributes in the Great Lakes–St. Lawrence mixedwood forests of southern Ontario, Canada. Inventory attribute estimates were validated at the stand level using independent reference data acquired for 27 intensively sampled stands. Top height, Lorey’s height, gross total volume for merchantable stems, merchantable stem volume, basal area, quadratic mean diameter, and total aboveground biomass were estimated with a relative RMSE of 13.52%, 7.24%, 14.61%, 16.27%, 14.42%, 12.25%, and 11.72%, respectively. Relative bias was < 1% for all attributes except top height (10.34%), merchantable volume (3.37%), and basal area (1.68%). Accuracy and bias varied by forest type and stand-level validation was important for assessing model performance in different stand conditions. SPL data can be used to generate accurate, area-based forest inventories in mixedwood forests that have a multitude of tree species and complex forest management histories.

Forest productivity is a crucial variable in forest planning, usually expressed as site index (SI). In Nordic commercial forest inventories, SI is commonly estimated by a combination of aerial imag...

Gully erosion is a worldwide problem of land degradation and water quality, and it is also frequent in Brazil. Typically, anthropic influence is the major driver of gully evolution. To study and monitor gullies it is necessary to use specific instruments and methods to obtain accurate information. The objective of this study was to use Terrestrial Laser Scanning (TLS) to create digital elevation model (DEM) accurately and define morphometric variables that characterize gullies in a mountainous relief. Two different interpolations were evaluated using the Topogrid and GridSurfaceCreate algorithms to elaborate DEM. Topographic profile for gullies was used to assess modeling quality. The DEM of the Gully 1 (G1) from the Topogrid algorithm estimated soil loss of 49%, whereas the GridSurfaceCreate algorithm estimated a soil loss of 97%, in a period of 1 year. The estimated soil loss for the Gully 2 (G2) was 14% from the Topogrid, and 8% from the GridSurfaceCreate algorithm. The GridSurfaceCreate algorithm underestimated the volume to area ratio for G2 due to a failure on interpolating a region of low point representativity. The Topogrid algorithm represented better the terrain irregularities, as observed through the topographic profiles traced in three regions of G1 and G2. Statistical analysis showed that the GridSurfaceCreate algorithm presented lower accuracy in estimating elevations. The underestimation trend of this algorithm was also observed in G2. The gullies showed considerable soil losses, which may reduce the areas suitable for agricultural activities, and silting up of water courses. The Topogrid algorithm presented satisfactory results, denoting great potential to produce morphometric data of gullies.

Airborne laser scanning (ALS) technology delivers large amount of data collected from airborne level. These data are used for many different applications in forestry, civil engineering, environmental studies and others. To acquire the best possible results from the data, accuracy analysis is a necessary part of data processing chain. Therefore, considering the increasing interest worldwide in the use of laser scanning data, improving the quality control (QC) tools is a crucial pursuit. This study underlines the possible error sources, summarises the existing QC knowledge for ALS data and proposes an optimised QC procedure. The procedure was implemented in selected applications and evaluated for three different environments, namely, forests, rural areas and croplands. The proposed solution is almost fully automatic outside from the module that supports the operator in the classification examination. The workflow is scalable and can be expanded with new modules that enhance the functionality. The presented procedures can save up to 30 min of manual checks for every 1 km2 area.

The objective of this study was to evaluate four airborne laser scanning derived indicators for monitoring tropical forest stands at different stages of sustainable management practices. The tradit...

The estimation of forest attributes at the local scale as well as in wall-to-wall approaches benefits from the integration of remote sensing data such as airborne laser scanning (ALS). A poor level...

Mobile mapping vehicles, equipped with cameras, laser scanners (in this paper referred to as light detection and ranging, LiDAR), and positioning systems are limited to acquiring surface data. However, in this paper, a method to fuse both LiDAR and 3D ground penetrating radar (GPR) data into consistent georeferenced point clouds is presented, allowing imaging both the surface and subsurface. Objects such as pipes, cables, and wall structures are made visible as point clouds by thresholding the GPR signal’s Hilbert envelope. The results are verified with existing utility maps. Varying soil conditions, clutter, and noise complicate a fully automatized approach. Topographic correction of the GPR data, by using the LiDAR data, ensures a consistent ground height. Moreover, this work shows that the LiDAR point cloud, as a reference, increases the interpretability of GPR data and allows measuring distances between above ground and subsurface structures.

Accurate assessment of forest structure and biomass is hampered by extensive field measurements that are time-consuming, costly, and inefficient. This is especially true in mangrove forests that have developed complex above-ground root structures for stability and survival in the harsh, anaerobic, and reducing conditions of water-logged sediments. These diverse structures can differ even among similar species, providing complex three dimensional structures and making them difficult to accurately assess using traditional allometric methods. Terrestrial laser scanners (TLS) have been used widely in collecting forest inventory information in recent years, mainly due to their fine-scale, detailed spatial measurements and rapid sampling. In this work we detected stems and roots in TLS data from three mangrove forests on Pohnpei Island in Micronesia using 3D classification techniques. After removing noise from the point cloud, the training set was acquired by filtering the facets of the point cloud based on angular orientation. However, many mangrove trees contain above-ground roots, which can incorrectly be classified as stems. We consequently trained a supporting classifier on the roots to detect omitted root returns (i.e., those classified as stems). Consistency was assessed by comparing TLS results to concurrent field measurements made in the same plots. The accuracy and precision for TLS stem classification was 82% and 77%, respectively. The same values for TLS root detection were 76% and 68%. Finally, we simulated the stems using alpha shapes for volume estimation. The average consistency of the TLS volume assessment was 85%. This was obtained by comparing the plot-level mean stem volume (m3/ha) between field and TLS data. Additionally, field-measured diameter-at-breast-heights (DBH) were compared to the lidar-derived DBH using the reconstructed stems, resulting in 74% average accuracy and an RMSE of 7.52 cm. This approach can be used for automatic structural evaluation, and could contribute to more accurate biomass assessment of complex mangrove forest environments as part of forest inventories or carbon stock assessments.

The use of Non-Destructive Testing (NDT) applied to construction materials allows to highlight and characterize their features, especially in the case of old buildings. The multi-technique high resolution 3D modelling described here is aimed to investigate the conservation state of the central column of a colonnade in the ancient church of Saints Lorenzo and Pancratio, dating to about the second half of the thirteenth century and located in the old town of Cagliari (Italy). This column was considered of interest because its longitudinal axis deviates from its ideal position and it appears the most deteriorated. In this work we describe the integrated application of 3D diagnostic methods, i.e. Terrestrial Laser Scanner (TLS), close range photogrammetry (CRP) and ultrasonic tomography supported by petrographic investigations. They were used to improve the diagnostic process of the conservation state of the investigated column. The TLS technique was supported by CRP to obtain a natural colour texturized 3D model of the column. The geometrical anomaly maps derived from the data of the TLS-CRP survey show the presence of some anomalies worthy of attention. Starting from the 3D reconstruction with previous techniques we planned and implemented a 3D ultrasonic tomography. Ultrasonic tomography proved to be a successful tool in identifying internal defects, as well as the presence of voids and flaws within the materials through the analysis of the propagation of ultrasonic waves. The integration of the three non-invasive techniques supported by petrographical analyses demonstrates its potential in reducing ambiguities since each technique brings its clue to the overall diagnostic process.

Since 2011 an intensive international cooperation has been in place between the Institute for Applied Photogrammetry and Geoinformatics (IAPG) at the Jade University of Applied Sciences in Oldenburg, Germany, and various Ukrainian universities. Following an initial contact by Prof. Gottfried Konecny, the first visit was organized, and was followed by many more. In subsequent years an intensive cooperation was established particularly with the National University for Construction and Architecture (KNUCA) in Kiev. In addition to architects and civil engineers, KNUCA also trains geodesists, geo-information scientists and landscape planners. The cooperation today includes the reciprocal exchange of scientists and students, research projects, courses and cooperation at many other levels. In addition, a commercial company has been established, SPM3D LLC, which now employs more than 14 engineers in the field of 3D acquisition, point cloud processing and modeling. This article summarizes the history of the cooperation and presents the results of associated student projects. In addition, results of joint work on the development of a virtual laser scanner are presented, part of a German-Ukrainian initiative to digitize teaching.

In this study, for five sites around the world, we look at the effects of different model types and variable selection approaches on forest yield modelling performances in an area-based approach (ABA). We compared ordinary least squares regression (OLS), k-nearest neighbours (kNN) and random forest (RF). Our objective was to test if there are systematic differences in accuracy between OLS, kNN and RF in ABA predictions of growing stock volume. The analyses are based on a 5-fold cross-validation at five study sites: an eucalyptus plantation, a temperate forest and three different boreal forests. Two completely independent validation datasets were also available for two of the boreal sites. For the kNN, we evaluated multiple measures of distance including Euclidean, Mahalanobis, most similar neighbour (MSN) and an RF-based distance metric. The variable selection approaches we examined included a heuristic approach (for OLS, kNN and RF), exhaustive search among all combinations (OLS only) and all variables together (RF only). Performances varied by model type and variable selection approaches among sites. OLS and RF had similar accuracies and were more efficient than any of the kNN variants. Variable selection did not affect RF performance. Heuristic and exhaustive variable selection performed similarly for OLS. kNN fared the poorest amongst model types, and kNN with RF distance was prone to overfitting when compared with a validation dataset. Additional caution is therefore required when building kNN models for volume prediction though ABA, being preferable instead to opt for models based on OLS with some variable selection, or RF with all variables together.

We present an approach for automatically detecting the positions of tree trunks, for determining their corresponding diameter at breast height (DBH), and for assessing the shape of tree trunks from 3D point clouds derived from unmanned aerial vehicle borne laser scanning (ULS). The experiments are carried out with point clouds from both a RIEGL miniVUX-1DL and from a RIEGL VUX-1UAV. The results reveal that the autonomous stem detection recognizes 91.0% and 77.6% of the stems, respectively, and that the DBH can be modeled with biases of 2.86 cm and 0.95 cm for 80.6% and 61.2% of the trees, when compared to field measurements. We further demonstrate that, compared to terrestrial laser scanning (TLS) data, the stem diameters along the tree can be estimated with biases below 3.4 cm and 1.4 cm for the two systems, respectively, up to a tree height of 12 m for stems with a DBH above 20 cm. Our experiments further reveal the accuracy of diameter estimations to be mainly dominated by the tree’s diameter with better accuracies for larger stems, while the completeness, with which a stem is covered by points, has little influence, as long as half of the stem circumference is captured. The absolute point count on the stem does not impact the estimation accuracy of all stem parameters, but is critical to the completeness with which a scene can be reconstructed. Conversely, we demonstrate the precision of the laser scanner to be a key factor for the accuracy of the stem diameter estimations, as in our experiments, we found the accuracies of the estimations from the VUX-1UAV to be higher than the ones from the miniVUX-1DL. The findings of our study assist to evaluate the potential of ULS for forest monitoring and management and allow for conclusions regarding the required point cloud qualities and, thus, the mission planning of ULS acquisitions, in order to deliver data products, which fulfill the requirements for an operational application in forest inventories.

Individual tree height (ITH) is one of the most important vertical structure parameters of a forest. Field measurement and laser scanning are very expensive for large forests. In this paper, we propose a cost-effective method to acquire ITHs in a forest using the optical overlapping images captured by an unmanned aerial vehicle (UAV). The data sets, including a point cloud, a digital surface model (DSM), and a digital orthorectified map (DOM), were produced from the UAV imagery. The canopy height model (CHM) was obtained by subtracting the digital elevation model (DEM) from the DSM removed of low vegetation. Object-based image analysis was used to extract individual tree crowns (ITCs) from the DOM, and ITHs were initially extracted by overlaying ITC outlines on the CHM. As the extracted ITHs were generally slightly shorter than the measured ITHs, a linear relationship was established between them. The final ITHs of the test site were retrieved by inputting extracted ITHs into the linear regression model. As a result, the coefficient of determination (R2), the root mean square error (RMSE), the mean absolute error (MAE), and the mean relative error (MRE) of the retrieved ITHs against the measured ITHs were 0.92, 1.08 m, 0.76 m, and 0.08, respectively.

Automatic extraction of road features from LiDAR data is a fundamental task for different applications, including asset management. The availability of updated and reliable models is even more important in the context of smart roads. One of the main advantages of LiDAR data compared with other sensing instruments is the possibility to directly get 3D information. However, the task of deriving road networks form LiDAR data acquired with Airborne Laser Scanning (ALS) may be quite complex due to occlusions, low feature separability and shadowing from contextual objects. Indeed, even if roads elements can be identified in the ALS point cloud, the automated identification of the network starting form them can be involved due to large variability in the size of roads, shapes and presence of connected off-road features such as parking lots. This paper presents a workflow aimed at partially solving the automatic creation of a road network from high-resolution ALS data. The presented method consists of three main steps: (i) labelling of road points; (ii) a multi-level voting scheme; and (iii) the regularization of the extracted road segments. The developed method has been tested using the “Vaihingen”, “Toronto” and “Tobermory” data set provided by the ISPRS.

Precise and accurate three-dimensional geospatial data has become increasingly available thanks to advances in both Terrestrial Laser Scanning (TLS) and Structure-from-Motion Photogrammetry (SfM). These tools provide valuable information for mapping geomorphological features and detect surface changes in mountainous environments. The exploitation of 3D point-clouds has been proven tremendously useful in the field of geosciences. It remains, however, controversial whether cost efficient photogrammetry can provide as accurate and reliable geospatial information as the significantly more expensive laser scanning or not. In this study, a rockfall case site in the territory of Obergurgl, Austria, is investigated in order to provide answers to the above question in a complex environment. The analysis includes different terrestrial photogrammetry configurations aiming to comprehensively define the strengths and limitations of terrestrial photogrammetry over TLS. The latter constitutes an optimized methodology that provides guidelines for costly future assessments as part of the site investigation phase in geohazard management. There are no doubts that compared to traditional and conventional surveying methods TLS and Photogrammetry both offer products much faster and with a much higher data density. In the current study, we show that when photogrammetry is applied following a well-defined optimized strategy, it can be potentially an adequate alternative to more costly TLS datasets for mass movement assessment and monitoring purposes.

Caves have been used by humans and animals for several thousand years until present but, at these time scales, their structures can rapidly change due to erosion and concretion processes. For this reason, the availability of precise 3D models improves the data quality and quantity allowing the reconstruction of their ancient appearance, structure and origin. However, caves are usually characterised by lack of light, high percentage of relative humidity, narrow spaces and complex morphology. Thus, quite often the traditional topographic instruments cannot be employed. In the La Sassa cave (Sonnino, Italy) a huge deposit ranging from Pleistocene to the Second World War has been found and stratigraphic evidence suggested that the shape of the cave and its entrance might have been different. In this paper, the fusion of the internal and external 3D photogrammetric models of the La Sassa, made to support the archaeological excavations, is presented, A Nikon camera with a fisheye lens and a smartphone camera have been used to survey the internal part of the cave, while an aerial drone has been employed for the external area. The two models have been georeferenced and scaled using GCPs acquired by a double frequency GNSS (GPS and GLONASS) receiver. A low-resolution DTM derived from a previous aerial laser scanning survey and the 3D models have been elaborated in CloudCompare environment to highlight the complete morphology of the cave and its surroundings.

This work presents a data acquisition framework and the technical details of a permanent terrestrial laser scanning (TLS) measurement station for high spatial and temporal resolution forest observation that was developed in the Finnish Geospatial Research Institute. The TLS measurement station was established to provide hyper-temporal time series of three-dimensional point cloud data for long term monitoring of a boreal forest. Time series data acquisition framework consists of regular 14-minute scans performed by a RIEGL VZ-2000i laser scanner in every 30 minutes, resulting in the collection of 48 scans per day. The entire framework includes the setting up of the laser scanner, the initialization of daily project, the scanning data acquisition over a preset time window, the storage management of the collected data at a local measurement computer, and the transfer of data from the measurement computer to network-attached storage (NAS) for further data processes. The operability of the proposed TLS measurement station was first piloted at a test area of about 32,500 m2 in Southern Finland (60°09'N, 24°32'E). A set of several long monitoring experiments were performed over the whole growing season from the beginning of April to the end of October in 2019. As preliminary results, the time series outputs have captured detailed information on the phenological changes in the test site with sub-centimetre accuracy. For instance, it was possible to visualize plant dynamics phenomena, such as the sprouting of leaves in spring and their falling in autumn.

Image matching of aerial or satellite images and Airborne Laser Scanning (ALS) are the two main techniques for the acquisition of geospatial information (3D point clouds), used for mapping and 3D modelling of large surface areas. While ALS point cloud classification is a widely investigated topic, there are fewer studies related to the image-derived point clouds, even less for point clouds derived from stereo satellite imagery. Therefore, the main focus of this contribution is a comparative analysis and evaluation of a supervised machine learning classification method that exploits the full 3D content of point clouds generated by dense image matching of tri-stereo Very High Resolution (VHR) satellite imagery. The images were collected with two different sensors (Pléiades and WorldView-3) at different timestamps for a study area covering a surface of 24 km2, located in Waldviertel, Lower Austria. In particular, we evaluate the performance and precision of the classifier by analysing the variation of the results obtained after multiple scenarios using different training and test data sets. The temporal difference of the two Pléiades acquisitions (7 days) allowed us to calculate the repeatability of the adopted machine learning algorithm for the classification. Additionally, we investigate how the different acquisition geometries (ground sample distance, viewing and convergence angles) influence the performance of classifying the satellite image-derived point clouds into five object classes: ground, trees, roads, buildings, and vehicles. Our experimental results indicate that, in overall the classifier performs very similar in all situations, with values for the F1-score between 0.63 and 0.65 and overall accuracies beyond 93%. As a measure of repeatability, stable classes such as buildings and roads show a variation below 3% for the F1-score between the two Pléiades acquisitions, proving the stability of the model.

Acquisition of three-dimensional (3D) structural information is significant for forest measurements. To achieve faster data collection in forests, we design a backpack laser scanning (BLS) system using a single mobile laser scanning (MLS) scanner and specific to forest environments. The simultaneous localization and mapping (SLAM) approach based on the natural geometric characteristics of trees is used for BLS-based forest mapping, in which the skeleton line of the individual tree is extracted for scan matching and the incremental maps are adopted for global optimization of all the BLS point clouds. The final experimental results show that the SLAM-based BLS system achieves accurate forest plots mapping and allows reaching low mapping errors, in which the mean errors are approximately 3 cm in the horizontal and 2 cm in the vertical direction.

In this paper we introduce the first dual-wavelength, kinematic backpack laser scanning system and its application on high resolution 3D terrain modelling of permafrost landforms. We discuss the data processing pipeline from acquisition to preparation, system calibration and terrain model process. Topographic information is vital for planning and monitoring tasks in urban planning, road construction for mass calculations, and mitigation of flood and wind related risks by structural design in coastal areas. 3D data gives possibility to understand natural processes inducing changes in the terrain, such as the cycles of thaw-freeze in permafrost regions. Through an application case on permafrost landforms in the Arctic we present the field practices and data processing applied, characterize the data output and discuss the precision and accuracy of the base station, trajectory and point cloud data. Two pulsed time of flight ranging, high performance mobile laser scanners were used in combination with a near navigation grade GNSS-IMU positioning on a kinematic backpack platform. The study shows that with a high-end system 15 mm absolute accuracy of 3D data could be achieved using PPP processing for the GNSS base station and multi-pass differential trajectory post-processing. The PPP solution shows millimetre level agreement (Easting 6 mm, Northing 4 mm, and elevation 8 mm standard deviations) for the base station coordinates over an 11 day period. The point cloud residual standard deviation for angular boresight misalignment was 27 mm. The absolute distance between ground surfaces from interactive analysis was 17 mm with 13 mm standard deviation (n = 64). The proposed backpack laser scanning provides accurate and precise 3D data and performance over considerable land surface area for detailed elevation modelling and analysis of the morphology of features of interest. The high density point cloud data permits fusion of the dual-wavelength lidar reflectance data into spectral products.

Automated recognition of terrain structures is a major research problem in many application areas. These structures can be investigated in raster products such as Digital Elevation Models (DEMs) generated from Airborne Laser Scanning (ALS) data. Following the success of deep learning and computer vision techniques on color images, researchers have focused on the application of such techniques in their respective fields. One example is detection of structures in DEM data. DEM data can be used to train deep learning models, but recently, Du et al. (2019) proposed a multi-modal deep learning approach (hereafter referred to as MM) proving that combination of geomorphological information help improve the performance of deep learning models. They reported that combining DEM, slope, and RGB-shaded relief gives the best result among other combinations consisting of curvature, flow accumulation, topographic wetness index, and grey-shaded relief. In this work, we approve and build on top of this approach. First, we use MM and show that combinations of other information such as sky view factors, (simple) local relief models, openness, and local dominance improve model performance even further. Secondly, based on the recently proposed HR-Net (Sun et al., 2019), we build a tinier, Multi-Modal High Resolution network called MM-HR, that outperforms MM. MM-HR learns with fewer parameters (4 millions), and gives an accuracy of 84:2 percent on ZISM50m data compared to 79:2 percent accuracy by MM which learns with more parameters (11 millions). On the dataset of archaeological mining structures from Harz, the top accuracy by MM-HR is 91:7 percent compared to 90:2 by MM.

The classification of Mobile Laser Scanner (MLS) data is challenging due to the combination of high variation in point density with a high variation of object appearances. The way how objects appear in the MLS data highly depends on the speed and orientation of the mobile mapping platform and the occlusion by other vehicles. There have been many approaches dealing with the geometric and contextual appearance of MLS points, voxels and segments to classify the MLS data. We present a completely different strategy by fusing the MLS data with a large scale topographic map. Underlying assumption is that the map delivers a clear hint on what to expect in the MLS data, at its approximate location. The approach presented here first fuses polygon objects, such as road, water, terrain and buildings, with ground and non-ground MLS points. Non-ground MLS points above roads and terrain are further classified by segmenting and matching the laser points to corresponding map point objects. The segmentation parameters depend on the class of the map points. We show that the fusion process is capable of classifying MLS data and detecting changes between the map and MLS data. The segmentation algorithm is not perfect, at some occasions not all the MLS points are correctly assigned to the corresponding map object. However, it is without doubt that the proposed map fusion delivers a very rich labelled point cloud automatically, which in future work can be used as training data in deep learning approaches.

Agricultural robotics rely on digital tools and sensor integration in order to improve efficiency and sustainability of cultivations. One part of orchard inventory is the identification of a tree trunk i.e. localization and diameter determination. However, this is a challenging task, due to thin trunks, presence of leaves and low branches. In this paper we present a case study for determining these parameters using the example of peach orchard, for which a high-density LiDAR data (over 3000 points/m2) was obtained with a small unmanned aerial system (UAS) during a leafy and leafless season. We applied point thresholding by height and by components of normal vector, in order to identify points reflected from trunks. Alpha-shape algorithm was used to aggregate together points, that belong to the same trunk and their centroid determined the trunk location. Trunk diameters were calculated using two alternative approaches: the Principal Component Analysis (PCA) and circle fit. For the leafy season trunk identification is challenging. Omission errors were caused due to few reflections from trunks and commission errors occurred because of the unfiltered reflections from low branches and young twigs oriented towards the ground. All 194 trunks were identified from data collected during the leafless season. The accuracy of tree location was 0.27 m and the accuracy of diameter determination using PCA was 0.03 m.

Terrestrial Laser scanner has been widely used in the field of forestry. Wood-leaf separation is the fundamental step to most applications of forestry. This paper presented a robust supervised learning method for wood and leaf classification by developing four new feature vectors. Fractal dimension is first calculated to indicate the difference of regularity or roughness between wood and leaf. Zenith angle and variation are presented to distinguish trunks or branches from leaves. The adaptive axis direction of cylinder is adopted to calculate the local point density precisely. Experimental results show that the supervised learning method using the four feature vectors presented in this paper can achieve a good classification performance. Both accuracy and F1 score are higher than the ones of the method using eigen value based feature vectors.

This paper presents the contribution of digital surveying techniques for the estimation of fluvial sediment transport. The aim is to create predictive models able to minimize or reduce the hydrogeological hazard, especially before or during critical meteorological events. The case study is the Caldone stream, a watercourse located in the municipality of Lecco (Italy). Structure-from-Motion photogrammetry and terrestrial laser scanning techniques were used to collect metric data about the morphology of the riverbed. Data acquisition was carried out to create a digital model of visible and submerged parts of the riverbed. Then, a second area with a sedimentation pool was selected to monitor the variation of the depth induced by progressive accumulation of sediments. As the pool is constantly covered by water, a low-cost bathymetric drone was used coupling the measured depth values with total station measurements to track the drone. Finally, the paper describes the implementation of an on-line data delivery platform able to facilitate retrieval of heterogeneous geospatial data, which are used in the developed numerical model for sediment transport. This service aims at providing simplified access to specific map layers without requiring knowledge about data formats and reference systems.

The technology of airborne laser scanning enables fast and accurate gathering spatial data containing also echoes from the terrain below the vegetation canopy that is beneficial for topographic mapping of wooded sandstone landscapes in Czechia, Poland, and Germany. The challengeable task is to determine the ground points in the point cloud because commonly used filtration methods do not successfully distinguish between vegetation and rock pillars and faces. In this paper, we replace filtration with classification approach using the features derived from characteristics of points within a neighbourhood of optimized sizes, such as eigenvalue-based features and echo ratio. Random Forest classifier is trained and tested on the manually labelled dataset with a density of almost 650 points/m2 from the Adršpach-Teplice Rocks. The overall accuracy reaches 87% but recall and precision of non-ground points are unsatisfactory. Misclassified non-ground points are located also within trees, thus we do not consider the result as suitable for DTM processing without further processing.

Efficient road edge extraction from point clouds acquired by Mobile Laser Scanning (MLS) is an important task because the road edge is one of the main elements of high definition maps. In this paper, we present a scanline-based road edge extraction method using a bend angle of scanlines from MLS point clouds. Scanline-based methods have advantages in that computational cost is low, it is easy to extract accurate road edges, and they are independent of driving speed of MLS compared to methods using unorganized points. In contrast, there are some problems with these methods where the extraction accuracy becomes low at curb cuts and intersections. The extraction accuracy becomes low caused by the scanning noise and small occlusion from weeds and fallen leaves. In addition, some parameters should be adjusted according to the mounting angle of the laser scanner on the vehicle. Therefore, we present a scanline-based road edge extraction method which can solve these problems. First, the points of the scanline are projected to a plane in order to reduce the influence of the mounting angle of the laser scanner on the vehicle. Next, the bend angle of each point is calculated by using filtered point clouds which are not vulnerable to small occlusions around the curb such as weeds. Then, points with a local maximum of bend angle and close to trajectories are extracted as seed points. Finally, road edges are generated by tracking based on bend angle of scanlines and smoothness of road edges from the seed points. In the experiments, our proposed methods achieved a completeness of over 95.3%, a correctness of over 95.0%, a quality of over 90.7%, and RMS difference less than 18.7 mm in total.

This paper proposes a methodology to calibrate the laser scanner of a Mobile Laser Scanning System (MLS) with the trajectory of the other MLS, both of which are installed directly above the top of both rails. Railway vehicle laser scanners systems of MLS are able to obtain 3D scanning map of the rail environment. In order to adapt the actual site condition of the maintenance works, we propose a calibration method with non-linear Least Mean Square calculation which use point clouds around poles along rails and sleepers of rails as cylindrical and planner constraints. The accuracy of 0.006 m between two laser point clouds can be achieved with this method. With the common planar and cylinder condition Leven-Marquardt method has been applied for this method. This method can execute without a good initial value for the extrinsic parameter and can shorten the processing time compared with the linear type of Least Mean Square method.

Unmanned aerial vehicle laser scanning (ULS) has recently become available for operational mapping and monitoring (e.g. for forestry applications or erosion studies). It combines advantages of terrestrial and airborne laser scanning, but there is still little proof of ULS accuracy. For the detection and monitoring of small-magnitude surfaces changes with multitemporal point clouds, an estimate of the level of detection (LOD) is required. The LOD is a threshold applied on distance measurements to separate real surface change (e.g. due to erosion or deposition by geomorphic processes) from errors. This paper investigates key components of the error budget for two ULS point clouds acquired for erosion monitoring at a grassland site in the Alps. In addition to the registration error and effects of the local surface roughness, we assess the positional uncertainties of each point that result from laser footprint effects, which are a function of the scanning geometry (including range, incidence angle and beam divergence). By removing erroneous points with an increasingly stricter point error criterion, we illustrate that the positional point errors strongly affect the LOD and discuss how this type of error can be mitigated. Moreover, our experimental results with three different surface classes (bare earth and rock, buildings and grassland) show that the level of detection tends to be slightly better for areas with bare earth and rock than for grass-covered areas (due to their roughness). For all these surface types reliable distance measurements are possible with sub-decimetre levels of detection.

Semantic segmentation is one of the main steps in the processing chain for Airborne Laser Scanning (ALS) point clouds, but it is also one of the most labour intensive steps, as it requires many labelled examples to train a classifier. National mapping agencies (NMAs) have to acquire nationwide ALS data every couple of years for their duties. Having point clouds cover different terrains such as flat or mountainous regions, a classifier often requires a refinement using additional data from those specific terrains. In this study, we present an algorithm, which is able to classify point clouds of similar terrain types without requiring any additional training data and which is still able to achieve overall F1-Scores of over 90% in most setups. Our algorithm uses up to two height distributions within a single cell in a rasterized point cloud. For each distribution, the empirical mean and standard deviation are calculated, which are the input for a Convolutional Neural Network (CNN) classifier. Consequently, our approach only requires the geometry of point clouds, which enables also the usage of the same network structure for point clouds from other sensor systems such as Dense Image Matching. Since the mean ground level varies with the observed area, we also examined five different normalisation methods for our input in order to reduce the ground influence on the point clouds and thus increase its transferability towards other datasets. We test our trained networks on four different tests sets with the classes’ ground, building, water, non-ground and bridge.

Terrestrial laser scanners are commonly used for remotely sensing natural surfaces into 3D point clouds. Time series of such 3D point clouds can be analysed to gain information of surface changes that are induced by Earth surface shaping processes. The atomic unit in time series analysis is a bitemporal change detection and quantification. This should involve an estimation of the minimum quantifiable change, the Level of Detection, to separate signal from noise, e.g. stemming from the measurement. To enable such an estimation through error propagation, a model of the sensing instrument’s measurement uncertainty is required. In this work, we present an investigation on the ranging component of terrestrial laser scanning on this uncertainty and its influence on 3D distances between point clouds of two epochs. Specifically, we analyse the effects of incidence angle, intensity and range for different object materials, and make additional considerations with respect to waveform information returned by the sensor. We estimate a model for the rangefinder uncertainty of a terrestrial laser scanner and apply it on experimental data. The results show that using a sensor-specific model of ranging uncertainty allows an appropriate estimation of the Level of Detection. At a range of 60 m and a rotational displacement of 10°, this Level of Detection ranges between 0.1 mm to 1 mm for a white and a grey surface and up to 5 mm for a black surface. The completeness of the detection of significant change ranges from 60.2 % (black) to 89.8 % (grey) for the proposed method and from 65.5 % to 88.9 % for the baseline, when compared to tachymeter measurements. The similarity between the results is expected and suggests the validity of error propagation for the derivation of the Level of Detection.

Background The aim of this study was to construct a nationwide stand age model by using National Forest Inventory (NFI) data and nationwide airborne laser scanning (ALS) data. In plantation forestry, age is usually known. While this is not the case in boreal managed forests, age is still seldom predicted in forest management inventories. Measuring age accurately in situ is also very laborious. On the other hand, tree age is one of the accurately measured sample tree attributes in NFI field data. Many countries also have a nationwide coverage of airborne laser scanning (ALS) data. In this study, we merged these data sources and constructed a nationwide, area-based model for stand age. Results While constructing the model, we omitted old forests from the data, since the correlation between ALS height metrics and stand age diminished at stands with age > 100 years. Additionally, the effect of growth conditions was considerable, so we also utilized different geographical and NFI variables such as site fertility and soil type in the modeling. The resultant nationwide model for the stand age of managed forests yielded a root mean square error (RMSE) of about 14 years. The model could be improved further by additional forest structure variables, but such information may not be available in practice. Conclusions The results showed that the prediction of stand age by ALS, geographical and NFI information was challenging, but still possible with moderate success. This study is an example of the joint use of NFI and nationwide ALS data and re-use of NFI data in research.

By utilizing the airborne laser scanning (ALS) and terrestrial laser scanning (TLS), the land surface information from both top view and side view can be captured rapidly. However, due to the different perspective views, resolutions, and ranges, the automatic localization of multiple TLS scans to ALS is challenging. To address this issue, this paper proposes a novel skyline context-based method. First, the ground points in ALS are extracted and used as potential TLS locations, and the corresponding skyline contexts are generated. After that, a 3D skyline-based k-d tree is built for searching the corresponding coarse localizations of TLS scans. The final refinement is done by the trimmed iterative closest point algorithm (T-ICP). 5 datasets with different ALS sizes and over one hundred TLS scans are undertaken to evaluate the performance of the proposed method. For one ALS data with mean point distance of 0.1 m, the average localization accuracy reached about 0.13 m. The experimental results indicate that the proposed method performs well for automatic localization of TLS scans to ALS point clouds, with advantages in both precision and adaptability.

A vehicle extraction method is proposed in this paper to extract vehicles in urban areas more accurately from airborne point clouds. First, the ground points are separated from the non-ground point...

The main objective of this study is the development of a simulation and optimization method for wide-area terrain mapping with terrestrial laser scanning (TLS). The problem can be stated as follows...

The combined use of unmanned aerial vehicles (UAVs) and structure-from-motion (SfM) is rapidly growing as a cost-effective alternative to airborne laser scanning (lidar) for reconstructing glacier surfaces. Here we present a thorough analysis of the precision and accuracy of a photogrammetric point cloud (PPC) constructed through SfM from UAV-acquired imagery over the spring snow surface at Haig Glacier, Alberta, Canada, the first of its kind in a glaciological setting. An aerial lidar survey conducted concurrently with UAV surveys was used to examine spatial patterns in the PPC accuracy. We found a median error in the PPC of −0.046 ± 0.067 m, with a 95% quantile of 0.218 m. Mean precision of the PPC was 0.199 m, with large spatially clustered outliers. We found an association between high-error, low-precision, and high-surface roughness in the PPC, likely due to illumination characteristics of the snow surface. Glacier surface reconstructions are important for geodetic mass balance measurements, giving key insights into changing climate where in situ measurements are difficult to obtain. The PPC errors are small enough that they would have minimal effects on total mass balance, should the technique be applied across the glacier.

International audience; Precise structural information collected from plots is significant in the management of and decision-making regarding forest resources. Currently, laser scanning is widely used in forestry inventories to acquire three-dimensional (3D) structural information. There are three main data-acquisition modes in ground-based forest measurements: single-scan terrestrial laser scanning (TLS), multi-scan TLS and multi-single-scan TLS. Nevertheless, each of these modes causes specific difficulties for forest measurements. Due to occlusion effects, the single-scan TLS mode provides scans for only one side of the tree. The multi-scan TLS mode overcomes occlusion problems, however, at the cost of longer acquisition times, more human labor and more effort in data preprocessing. The multi-single-scan TLS mode decreases the workload and occlusion effects but lacks the complete 3D reconstruction of forests. These problems in TLS methods are largely avoided with mobile laser scanning (MLS); however, the geometrical peculiarity of forests (e.g., similarity between tree shapes, placements, and occlusion) complicates the motion estimation and reduces mapping accuracy.Therefore, this paper proposes a novel method combining single-scan TLS and MLS for forest 3D data acquisition. We use single-scan TLS data as a reference, onto which we register MLS point clouds, so they fill in the omission of the single-scan TLS data. To register MLS point clouds on the reference, we extract virtual feature points that are sampling the centerlines of tree stems and propose a new optimization-based registration framework. In contrast to previous MLS-based studies, the proposed method sufficiently exploits the natural geometric characteristics of trees. We demonstrate the effectiveness, robustness, and accuracy of the proposed method on three datasets, from which we extract structural information. The experimental results show that the omission of tree stem data caused by one scan can be compensated for by the MLS data, and the time of the field measurement is much less than that of the multi-scan TLS mode. In addition, single-scan TLS data provide strong global constraints for MLS-based forest mapping, which allows low mapping errors to be achieved, e.g., less than 2.0 cm mean errors in both the horizontal and vertical directions.

This study had two main aims: (1) to provide a comprehensive review of terrestrial laser scanner (TLS) point cloud registration methods and a better understanding of their strengths and weaknesses; and (2) to provide a large-scale benchmark data set (Wuhan University TLS: Whu-TLS) to support the development of cutting-edge TLS point cloud registration methods, especially deep learning-based methods. In particular, we first conducted a thorough review of TLS point cloud registration methods in terms of pairwise coarse registration, pairwise fine registration, and multiview registration, as well as analyzing their strengths, weaknesses, and future research trends. We then reviewed the existing benchmark data sets (e.g., ETH Dataset and Robotic 3D Scanning Repository) for TLS point cloud registration and summarized their limitations. Finally, a new benchmark data set was assembled from 11 different environments (i.e., subway station, high-speed railway platform, mountain, forest, park, campus, residence, riverbank, heritage building, underground excavation, and tunnel environments) with variations in the point density, clutter, and occlusion. In addition, we summarized future research trends in this area, including auxiliary data-guided registration, deep learning-based registration, and multi-temporal point cloud registration.