Non-ground Points Research Articles

Spatial Analysis Using Temporal Point Clouds in Advanced GIS: Methods for Ground Elevation Extraction in Slant Areas and Building Classifications

Deriving 3D urban development patterns is necessary for urban planners to control the future directions of 3D urban growth considering the availability of infrastructure or being prepared for fundamental infrastructure. Urban metrics have been used so far for quantification of landscape and land-use change. However, these studies focus on the horizontal development of urban form. Therefore, questions remain about 3D growth patterns. Both 3D data and appropriate 3D metrics are fundamentally required for vertical development pattern extraction. Airborne light detection and ranging (Lidar) as an advanced remote-sensing technology provides 3D data required for such studies. Processing of airborne lidar to extract buildings’ heights above a footprint is a major task and current automatic algorithms fail to extract such information on vast urban areas especially in hilly sites. This research focuses on proposing new methods of extraction of ground points in hilly urban areas using autocorrelation-based algorithms. The ground points then would be used for digital elevation model generation and elimination of ground elevation from classified buildings points elevation. Technical novelties in our experimentation lie in choosing a different window direction and also contour lines for the slant area, and applying moving windows and iterating non-ground extraction. The results are validated through calculation of skewness and kurtosis values. The results show that changing the shape of windows and their direction to be narrow long squares parallel to the ground contour lines, respectively, improves the results of classification in slant areas. Four parameters, namely window size, window shape, window direction and cell size are empirically chosen in order to improve initial digital elevation model (DEM) creation, enhancement of the initial DEM, classification of non-ground points and final creation of a normalised digital surface model (NDSM). The results of these enhanced algorithms are robust for generating reliable DEMs and separation of ground and non-ground points in slant urban scenes as evidenced by the results of skewness and kurtosis. Offering the possibility of monitoring urban growth over time with higher accuracy and more reliable information, this work could contribute in drawing the future directions of 3D urban growth for a smarter urban growth in the Smart Cities paradigm.

An Improved Skewness Balancing Filtering Algorithm Based on Thin Plate Spline Interpolation

Most filtering algorithms suffer from complex parameter settings or threshold adjusting. To solve this problem, this paper proposes an improved skewness balancing filtering algorithm based on thin plate spline (TPS) interpolation. The proposed algorithm filters the nonground points in an iterative manner. A reference surface that reflects the fluctuation of the terrain is generated using the TPS interpolation method. Accordingly, the elevation difference from each point to the surface can be calculated. By applying the skewness balancing principle to these elevation differences, nonground points can be removed automatically. To verify the validity and robustness of the proposed method, the datasets provided by the International Society for Photogrammetry and Remote Sensing (ISPRS) were adopted. The experimental results show that this presented method can adapt to complex environments and achieve a higher filtering accuracy than the traditional skewness balancing algorithm. Moreover, in comparison with the other eight filtering methods tested by the ISPRS and four improved filtering methods proposed recently, the proposed method achieved an average total error of 5.39%, which is smaller than that of most of these other methods.

Aboveground Tree Biomass Estimation of Sparse Subalpine Coniferous Forest with UAV Oblique Photography

In tree Aboveground Biomass (AGB) estimation, the traditional harvest method is accurate but unsuitable for a large-scale forest. The airborne Light Detection And Ranging (LiDAR) is superior in obtaining the point cloud data of a dense forest and extracting tree heights for AGB estimation. However, the LiDAR has limitations such as high cost, low efficiency, and complicated operations. Alternatively, the overlapping oblique photographs taken by an Unmanned Aerial Vehicle (UAV)-loaded digital camera can also generate point cloud data using the Aerial Triangulation (AT) method. However, limited by the relatively poor penetrating capacity of natural light, the photographs captured by the digital camera on a UAV are more suitable for obtaining the point cloud data of a relatively sparse forest. In this paper, an electric fixed-wing UAV loaded with a digital camera was employed to take oblique photographs of a sparse subalpine coniferous forest in the source region of the Minjiang River. Based on point cloud data obtained from the overlapping photographs, a Digital Terrain Model (DTM) was generated by filtering non-ground points along with the acquisition of a Digital Surface Model (DSM) of Minjiang fir trees by eliminating subalpine shrubs and meadows. Individual tree heights were extracted by overlaying individual tree outlines on Canopy Height Model (CHM) data computed by subtracting the Digital Elevation Model (DEM) from the rasterized DSM. The allometric equation with tree height (H) as the predictor variable was established by fitting measured tree heights with tree AGBs, which were estimated using the allometric equation on H and Diameter at Breast Height (DBH) in sample tree plots. Finally, the AGBs of all of the trees in the test site were determined by inputting extracted individual tree heights into the established allometric equation. In accuracy assessment, the coefficient of determination (R2) and Root Mean Square Error (RMSE) of extracted individual tree heights were 0.92 and 1.77 m, and the R2 and RMSE of the estimated AGBs of individual trees were 0.96 and 54.90 kg. The results demonstrated the feasibility and effectiveness of applying UAV-acquired oblique optical photographs to the tree AGB estimation of sparse subalpine coniferous forests.

Automatic DTM extraction from airborne LiDAR based on expectation-maximization

Filtering of ground points is a key step for most applications of airborne LiDAR point clouds. Although many filtering algorithms have been proposed in recent years, most of them suffer from parameter setting or thresholds fine-tuning. This is most often time-consuming and reduces the degree of automation of the applied algorithm. To overcome such problems, this paper proposes a threshold-free filtering algorithm based on expectation–maximization (EM). The filter is developed based on the assumption that point clouds are seen as a mixture of Gaussian models. Thus, the separation of ground points and non-ground points from point clouds is partitioning of the point clouds by a mixed Gaussian model that is used for screening ground points. EM is applied to realize the separation, which calculates the maximum likelihood estimates of the mixture parameters. Using the estimated parameters, the likelihoods of each point belonging to ground or non-ground are computed. Noticeably, point clouds are labeled as the component with a larger likelihood. The proposed method has been tested using the standard filtering datasets provided by the ISPRS. Experimental results showed that the proposed method performed the best in comparison with the classic progressive triangulated irregular network densification (PTD) and segment-based PTD methods in terms of omission error. The average omission error of the proposed method was 52.81% and 16.78% lower than the classic PTD method and the segment-based PTD method, respectively. Moreover, the proposed method was able to reduce its average total error by 31.95% compared to the classic PTD method.

Automated detection and decomposition of railway tunnels from Mobile Laser Scanning Datasets

Abstract Since the mid-19th century, the railway network has occupied a crucial place at the heart of the world's transport systems. Its infrastructure is often situated in harsh environments where an extreme event, or even daily use, could lead to a catastrophic accident. This is one of the main reasons why inspecting these constructions is so important. Despite the advances in this field, the human component continues to be part of the final inspection process. In order to improve on this, this paper shows the use of laser scanning as a leading technology in automating the inspection of railway infrastructures. The proposed methodologies provide the essential processed and classified data needed for the structural health monitoring of the various assets related to railways. It is divided into three main parts, which pre-process the point cloud, divide the cloud into ground and non-ground points, and detect the elements present in each of these clouds. The methods are validated in three case studies, each containing different railway tunnels. The results demonstrate that laser scanning technology, together with customized processing tools, can provide data for further structural operations with no requirement for either training in geomatics or high-performance computers for the data processing. Significant results are obtained for the developed classification methods: the classification of the tunnel elements returns a global F-Score of between 71 and 99% in a point-by-point comparison. With regard to the labelled rails classification, a global F-Score of 100% is achieved for the analyzed datasets, and between 56 and 73% for the point-by-point classification.

Automatic detection of building roofs from point clouds produced by the dense image matching technique

ABSTRACTThree-dimensional (3D) spatial information of object points is a vital requirement for many disciplines. Laser scanning technology and techniques based on image matching have been used extensively to produce 3D dense point clouds. These data are used frequently in various applications, such as the generation of digital surface model (DSM)/digital terrain model (DTM), extracting objects (e.g., buildings, trees, and roads), 3D modelling, and detecting changes. The aim of this study was to extract the building roof points automatically from the 3D point cloud data created via the image matching techniques with optical aerial images (with red, green, and blue band (RGB) and infrared (IR)). In the first stage of the study, as an alternative to laser scanning technology, which is more expensive than optical imaging systems, the 3D point clouds were produced by matching high-resolution images using a Semi Global Matching algorithm. The normalized difference vegetation index (NDVI) values for each point were calculated using the spectral information (RGB + IR) in the 3D point cloud data, and the points that represented the vegetation cover were determined using these values. In the second stage, existing ground and non-ground points that were free of vegetation cover were determined within the point cloud. Subsequently, only the points on the roof of the building were detected automatically using the proposed algorithm. Thus, points of the roofs of buildings located in areas with different topographic characteristics were detected automatically detected using only images. It was determined that the average values of correctness (Corr), completeness (Comp), and quality (Q) of the pixel-based accuracy analysis metrics were 95%, 98%, and 93%, respectively, in the selected test areas. According to the results of the accuracy analysis, it is clear that the proposed algorithm is very successful in automatic extraction of building roof points.

Analysis of Airborne LiDAR Point Clouds With Spectral Graph Filtering

Separation of ground and nonground measurements is an essential task in the analysis of light detection and ranging (LiDAR) point clouds; however, it is challenge to implement a LiDAR filtering algorithm that integrates the mathematical definition of various landforms. In this letter, we propose a novel LiDAR filtering algorithm that adapts to the irregular structure and 3-D geometry of LiDAR point clouds. We exploit weighted graph representations to analyze the 3-D point cloud on its original domain. Then, we consider airborne LiDAR data as an irregular elevation signal residing on graph vertices. Based on a spectral graph approach, we introduce a new filtering algorithm that distinguishes ground and nonground points in terms of their spectral characteristics. Our complete filtering framework consists of outlier removal, iterative graph signal filtering, and erosion steps. Experimental results indicate that the proposed framework achieves a good accuracy on the scenes with data gaps and classifies the nonground points on bridges and complex shapes satisfactorily, while those are usually not handled well by the state-of-the-art filtering methods.

A Voxel-Based Method for Automated Detection and Mapping of Light Poles on Rural Highways using LiDAR Data

The number of light poles and their position (in terms of density and offset off the roadside) have significant impacts on the safe operation of highways. In current practice, inventory of such information is performed in periodic site visits, which are tedious and time consuming. This makes inventory and health monitoring of poles at a network level extremely challenging. To relieve the burden associated with manual inventory of poles, this paper proposes a novel algorithm which can automatically obtain such information from remotely sensing data. The proposed algorithm works by first tiling point cloud data collected using light detection and ranging (LiDAR) technology into manageable data tiles of fixed dimensions. The data are voxelized and attributes for each data voxel are calculated to classify them into ground and nonground points. Connected components labeling is then used to perform 3D clustering of the data voxels. Further clustering is performed using a density-based clustering to combine connected components of the same object. The final step involves classifying different objects into poles and non-poles based on a set of decision rules related to the geometric properties of the clusters. The proposed algorithm was tested on a 4 km rural highway segment in Alberta, Canada, which had substantial variation in its vertical alignment. The algorithm was accurate in detecting nonground objects, including poles. Moreover, the results also highlight the importance of considering the length of the highway and its terrain when detecting nonground objects from LiDAR.

DTM generation from the point cloud using a progressive geodesic morphology and a modified Particle Swarm Optimization

ABSTRACTThe digital terrain model (DTM) is an important geospatial product used in many applications of geo-information systems. Advances in light detection and ranging technology leads to generate a dense and accurate point cloud from the Earth’s surface. Because of the existence of various types of non-ground object in the complex terrain, filtering of the point cloud data and DTM interpolation is still a challenging topic. In this article, a progressive geodesic morphology has been proposed to filter the non-ground points from the point cloud. In this regard, the noises were eliminated and the point cloud was gridded and rasterized. Then, an iterative procedure was carried out based on an elevation vector constructed by considering the terrain topography. The geodesic dilation and opening operators were applied in each iteration and some pixels were detected as the non-ground points. These pixels were labelled by the connected component analysis and the generated parcels were investigated using some geometric and structural features such as the object height, the slope of the terrain, the slope of the object boundary, as well as the boundary factor in order to decide about the considered parcel entity. Afterwards, the non-ground points were removed and the DTM was interpolated using the bare Earth points. A modified Particle Swarm Optimization (PSO) has been proposed to optimize the coefficients of the polynomial interpolation for generating a DTM with high accuracy. Results showed that the error of the proposed progressive geodesic morphology was on average 3.42% using an optimized set of parameters and 4.00% using a single set of parameters, respectively. These results represented the stability of the proposed filtering method against the predefined thresholds and this is an important factor to make it a practical method. In comparing the proposed modified PSO with the other interpolation algorithms, the evaluation results indicated that the proposed modified PSO interpolated data only with 0.0884 cm error, averagely. This means the proposed modified PSO has a significant performance in generating DTM.

FILTERING PHOTOGRAMMETRIC POINT CLOUDS USING STANDARD LIDAR FILTERS TOWARDS DTM GENERATION

Abstract. Digital Terrain Models (DTMs) can be generated from point clouds acquired by laser scanning or photogrammetric dense matching. During the last two decades, much effort has been paid to developing robust filtering algorithms for the airborne laser scanning (ALS) data. With the point cloud quality from dense image matching (DIM) getting better and better, the research question that arises is whether those standard Lidar filters can be used to filter photogrammetric point clouds as well. Experiments are implemented to filter two dense matching point clouds with different noise levels. Results show that the standard Lidar filter is robust to random noise. However, artefacts and blunders in the DIM points often appear due to low contrast or poor texture in the images. Filtering will be erroneous in these locations. Filtering the DIM points pre-processed by a ranking filter will bring higher Type II error (i.e. non-ground points actually labelled as ground points) but much lower Type I error (i.e. bare ground points labelled as non-ground points). Finally, the potential DTM accuracy that can be achieved by DIM points is evaluated. Two DIM point clouds derived by Pix4Dmapper and SURE are compared. On grassland dense matching generates points higher than the true terrain surface, which will result in incorrectly elevated DTMs. The application of the ranking filter leads to a reduced bias in the DTM height, but a slightly increased noise level.

A top-down strategy for buildings extraction from complex urban scenes using airborne LiDAR point clouds

In the field of airborne LiDAR point clouds processing, the extraction of buildings has been an active research area for many years. However, it is still difficult to distinguish buildings from vegetation and other objects by using the point entity or the segment entity in various complex urban scenes. Therefore, this paper proposes a simple and novel method using a top-down strategy based on the object entity, which takes an object and its surrounding points as a unit for analysis. Firstly, ground points are separated from non-ground points, and non-ground points are segmented for the detection of smooth regions. Secondly, the top-level processing recognizes the building regions from smooth regions based on their geometric and penetrating features. Finally, the down-level processing is employed to remove non-building points around buildings from each building region using topological, geometric and penetrating features. The ISPRS benchmark dataset is selected to perform the experiment, the result is compared with the state-of-the-art methods, and parameter sensitivity analysis is performed to evaluate the robustness of the proposed method. The evaluation result shows that the proposed method achieves good performance with respect to area-based and object-based quality, and it is robust when parameters are within reasonable ranges. Furthermore, the proposed method is utilized to process one large-scale dataset of Toronto. The result shows that the proposed method achieves a completeness of 96.2%, a correctness of 96.8% and a quality of 93.2% at the object level.

APPLICABILITY ANALYSIS OF CLOTH SIMULATION FILTERING ALGORITHM FOR MOBILE LIDAR POINT CLOUD

Abstract. Classifying the original point clouds into ground and non-ground points is a key step in LiDAR (light detection and ranging) data post-processing. Cloth simulation filtering (CSF) algorithm, which based on a physical process, has been validated to be an accurate, automatic and easy-to-use algorithm for airborne LiDAR point cloud. As a new technique of three-dimensional data collection, the mobile laser scanning (MLS) has been gradually applied in various fields, such as reconstruction of digital terrain models (DTM), 3D building modeling and forest inventory and management. Compared with airborne LiDAR point cloud, there are some different features (such as point density feature, distribution feature and complexity feature) for mobile LiDAR point cloud. Some filtering algorithms for airborne LiDAR data were directly used in mobile LiDAR point cloud, but it did not give satisfactory results. In this paper, we explore the ability of the CSF algorithm for mobile LiDAR point cloud. Three samples with different shape of the terrain are selected to test the performance of this algorithm, which respectively yields total errors of 0.44 %, 0.77 % and1.20 %. Additionally, large area dataset is also tested to further validate the effectiveness of this algorithm, and results show that it can quickly and accurately separate point clouds into ground and non-ground points. In summary, this algorithm is efficient and reliable for mobile LiDAR point cloud.

A THRESHOLD-FREE FILTERING ALGORITHM FOR AIRBORNE LIDAR POINT CLOUDS BASED ON EXPECTATION-MAXIMIZATION

Abstract. Filtering is a key step for most applications of airborne LiDAR point clouds. Although lots of filtering algorithms have been put forward in recent years, most of them suffer from parameters setting or thresholds adjusting, which will be time-consuming and reduce the degree of automation of the algorithm. To overcome this problem, this paper proposed a threshold-free filtering algorithm based on expectation-maximization. The proposed algorithm is developed based on an assumption that point clouds are seen as a mixture of Gaussian models. The separation of ground points and non-ground points from point clouds can be replaced as a separation of a mixed Gaussian model. Expectation-maximization (EM) is applied for realizing the separation. EM is used to calculate maximum likelihood estimates of the mixture parameters. Using the estimated parameters, the likelihoods of each point belonging to ground or object can be computed. After several iterations, point clouds can be labelled as the component with a larger likelihood. Furthermore, intensity information was also utilized to optimize the filtering results acquired using the EM method. The proposed algorithm was tested using two different datasets used in practice. Experimental results showed that the proposed method can filter non-ground points effectively. To quantitatively evaluate the proposed method, this paper adopted the dataset provided by the ISPRS for the test. The proposed algorithm can obtain a 4.48 % total error which is much lower than most of the eight classical filtering algorithms reported by the ISPRS.

CROSS VALIDATION ON THE EQUALITY OF UAV-BASED AND CONTOUR-BASED DEMS

Abstract. Unmanned Aerial Vehicles (UAV) have been widely used for Digital Elevation Model (DEM) generation in geographic applications. This paper proposes a novel framework of generating DEM from UAV images. It starts with the generation of the point clouds by image matching, where the flight control data are used as reference for searching for the corresponding images, leading to a significant time saving. Besides, a set of ground control points (GCP) obtained from field surveying are used to transform the point clouds to the user’s coordinate system. Following that, we use a multi-feature based supervised classification method for discriminating non-ground points from ground ones. In the end, we generate DEM by constructing triangular irregular networks and rasterization. The experiments are conducted in the east of Jilin province in China, which has been suffered from soil erosion for several years. The quality of UAV based DEM (UAV-DEM) is compared with that generated from contour interpolation (Contour-DEM). The comparison shows a higher resolution, as well as higher accuracy of UAV-DEMs, which contains more geographic information. In addition, the RMSE errors of the UAV-DEMs generated from point clouds with and without GCPs are ±0.5 m and ±20 m, respectively.

A simple terrain relief index for tuning slope-related parameters of LiDAR ground filtering algorithms

Ground filtering is an essential procedure in almost all LiDAR applications. However, most existing ground filtering algorithms require different amounts of user input to manually set up initial parameters, such as terrain relief amplitude and average slope, which is subjective, time consuming, and prone to errors. Here, we propose a simple terrain relief index derived from raw airborne LiDAR data to automatically tune the slope-related parameters of ground filtering algorithms. The terrain relief index is a ratio between the height difference of the entire point cloud and the maximum above ground level of non-ground points. The latter variable can be estimated with the maximum local height difference of raw LiDAR data through gridding. We validated our method using the benchmark airborne LiDAR datasets provided by the International Society for Photogrammetry and Remote Sensing. The results showed a high correlation (r = 0.876) between the terrain relief index and the referential terrain relief amplitude. The degree of correlation was greater across larger areas (r = 0.926) than small areas (r = 0.861) regardless of the type of land cover (e.g., city or forest). The terrain relief index was introduced into two existing filtering algorithms: Cloth Simulation Filtering (CSF) and Progressive Morphological (PM) Filter, by relating the terrain relief index to the cloth rigidness of the CSF and the slope threshold of the PM filter. To compare the results, the two algorithms were implemented both with manually tuned parameters and with the parameters derived from the terrain relief index. The results showed that there was only a slight discrepancy in average Total Error (0.1%) between them in the CSF, which means that the terrain relief index can automatically determine the cloth rigidness without noticeable loss of accuracy. The average difference between the slope threshold provided by the terrain relief index and the manually tuned optimal slope threshold was 0.142 rad (8.136°) for the PM filter, which is acceptable relative to manually parameter setting without any prior knowledge. The terrain relief index can estimate the terrain relief amplitude from raw airborne LiDAR data, and the parameter settings suggested in this paper for the filtering algorithm can improve automation.

Multimedia System for Real-Time Photorealistic Nonground Modeling of 3D Dynamic Environment for Remote Control System

Nowadays, unmanned ground vehicles (UGVs) are widely used for many applications. UGVs have sensors including multi-channel laser sensors, two-dimensional (2D) cameras, Global Positioning System receivers, and inertial measurement units (GPS–IMU). Multi-channel laser sensors and 2D cameras are installed to collect information regarding the environment surrounding the vehicle. Moreover, the GPS–IMU system is used to determine the position, acceleration, and velocity of the vehicle. This paper proposes a fast and effective method for modeling nonground scenes using multiple types of sensor data captured through a remote-controlled robot. The multi-channel laser sensor returns a point cloud in each frame. We separated the point clouds into ground and nonground areas before modeling the three-dimensional (3D) scenes. The ground part was used to create a dynamic triangular mesh based on the height map and vehicle position. The modeling of nonground parts in dynamic environments including moving objects is more challenging than modeling of ground parts. In the first step, we applied our object segmentation algorithm to divide nonground points into separate objects. Next, an object tracking algorithm was implemented to detect dynamic objects. Subsequently, nonground objects other than large dynamic ones, such as cars, were separated into two groups: surface objects and non-surface objects. We employed colored particles to model the non-surface objects. To model the surface and large dynamic objects, we used two dynamic projection panels to generate 3D meshes. In addition, we applied two processes to optimize the modeling result. First, we removed any trace of the moving objects, and collected the points on the dynamic objects in previous frames. Next, these points were merged with the nonground points in the current frame. We also applied slide window and near point projection techniques to fill the holes in the meshes. Finally, we applied texture mapping using 2D images captured using three cameras installed in the front of the robot. The results of the experiments prove that our nonground modeling method can be used to model photorealistic and real-time 3D scenes around a remote-controlled robot.

Joint Clusters and Iterative Graph Cuts for ALS Point Cloud Filtering

A novel method that combines joint clusters and iterative graph cuts for ALS point cloud filtering is proposed in this paper. The method first extracts clusters of points from the raw point cloud, and then classifies ground points at the cluster level. There are four main steps, i.e., two-step point cloud clustering, critical point extraction, initial terrain determination, and terrain densification based on iterative graph cuts. Smooth clusters, rough clusters, and scattered points are extracted by the two-step clustering to depict the raw point cloud, which reduces the complexity of raw data and the judgment difficulty in the subsequent procedures. Critical points of each cluster are extracted, and the initial terrain is determined among the smooth clusters. Using the initial terrain and critical points, iterative graph cuts is performed to segment ground and nonground points at the cluster level. Experiments on ISPRS dataset with a low point density and Utah dataset with a moderate point density show that our approach provides a satisfactory trade off between Type I and Type II errors. Moreover, our method significantly outperforms progressive TIN densification based filters, and successfully controls Type II errors.

A New Recursive Filtering Method of Terrestrial Laser Scanning Data to Preserve Ground Surface Information in Steep-Slope Areas

Landslides are one of the critical natural hazards that cause human, infrastructure, and economic losses. Risk of catastrophic losses due to landslides is significant given sprawled urban development near steep slopes and the increasing proximity of large populations to hilly areas. For reducing these losses, a high-resolution digital terrain model (DTM) is an essential piece of data for a qualitative or a quantitative investigation of slopes that may lead to landslides. Data acquired by a terrestrial laser scanning (TLS), called a point cloud, has been widely used to generate a DTM, since a TLS is appropriate for detecting small- to large-scale ground features on steep slopes. For an accurate DTM, TLS data should be filtered to remove non-ground points, but most current algorithms for extracting ground points from a point cloud have been developed for airborne laser scanning (ALS) data and not TLS data. Moreover, it is a challenging task to generate an accurate DTM from a steep-slope area by using existing algorithms. For these reasons, we developed an algorithm to automatically extract only ground points from the point clouds of steep terrains. Our methodology is focused on TLS datasets and utilizes the adaptive principal component analysis–triangular irregular network (PCA-TIN) approach. Our method was applied to two test areas and the results showed that the algorithm can cope well with steep slopes, giving an accurate surface model compared to conventional algorithms. Total accuracy values of the generated DTMs in the form of root mean squared errors are 1.84 cm and 2.13 cm over the areas of 5252 m2 and 1378 m2, respectively. The slope-based adaptive PCA-TIN method demonstrates great potential for TLS-derived DTM construction in steep-slope landscapes.

A NEW MASK FOR AUTOMATIC BUILDING DETECTION FROM HIGH DENSITY POINT CLOUD DATA AND MULTISPECTRAL IMAGERY

Abstract. In complex urban and residential areas, there are buildings which are not only connected with and/or close to one another but also partially occluded by their surrounding vegetation. Moreover, there may be buildings whose roofs are made of transparent materials. In transparent buildings, there are point returns from both the ground (or materials inside the buildings) and the rooftop. These issues confuse the previously proposed building masks which are generated from either ground points or non-ground points. The normalised digital surface model (nDSM) is generated from the non-ground points and usually it is hard to find individual buildings and trees using the nDSM. In contrast, the primary building mask is produced using the ground points, thereby it misses the transparent rooftops. This paper proposes a new building mask based on the non-ground points. The dominant directions of non-ground lines extracted from the multispectral imagery are estimated. A dummy grid with the target mask resolution is rotated at each dominant direction to obtain the corresponding height values from the non-ground points. Three sub-masks are then generated from the height grid by estimating the gradient function. Two of these sub-masks capture planar surfaces whose height remain constant in along and across the dominant direction, respectively. The third sub-mask contains only the flat surfaces where the height (ideally) remains constant in all directions. All the sub-masks generated in all estimated dominant directions are combined to produce the candidate building mask. Although the application of the gradient function helps in removal of most of the vegetation, the final building mask is obtained through removal of planar vegetation, if any, and tiny isolated false candidates. Experimental results on three Australian data sets show that the proposed method can successfully remove vegetation, thereby separate buildings from occluding vegetation and detect buildings with transparent roof materials. While compared to existing building detection techniques, the proposed technique offers higher objectbased completeness, correctness and quality, specially in complex scenes with aforementioned issues. It is not only capable of detecting transparent buildings, but also small garden sheds which are sometimes as small as 5 m2 in area.

CLASSIFIER-FREE DETECTION OF POWER LINE PYLONS FROM POINT CLOUD DATA

Abstract. High density airborne point cloud data has become an important means for modelling and maintenance of a power line corridor. Since, the amount of data in a dense point cloud is huge even in a small area, an automatic detection of pylons in the corridor can be a prerequisite for efficient and effective extraction of wires in a subsequent step. However, the existing solutions mostly overlook this important requirement by processing the whole data into one go, which nonetheless will hinder their applications to large areas. This paper presents a new pylon detection technique from point cloud data. First, the input point cloud is divided into ground and nonground points. The non-ground points within a specific low height region are used to generate a pylon mask, where pylons are found stand-alone, not connected with any wires. The candidate pylons are obtained using a connected component analysis in the mask, followed by a removal of trees by comparing area, shape and symmetry properties of trees and pylons. Finally, the parallelism property of wires with the line connecting pair of candidate pylons is exploited to remove trees that have the same area and shape properties as pylons. Experimental results show that the proposed technique provides a high pylon detection rate in terms of completeness (100 %) and correctness (100 %).