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

2018 ◽  
Vol XLII-3 ◽  
pp. 607-610 ◽  
Author(s):  
Z. Hui ◽  
P. Cheng ◽  
Y. Y. Ziggah ◽  
Y. Nie
Keyword(s):  
Expectation Maximization ◽  
Total Error ◽  
Gaussian Model ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Maximum Likelihood Estimates ◽  
Filtering Algorithms ◽  
Filtering Algorithm ◽  
Intensity Information ◽  
Em Method

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.

Development on Filtering Algorithms of Airborne LiDAR Point Clouds

2012 ◽  
Vol 226-228 ◽  
pp. 1892-1898
Author(s):  
Jian Qing Shi ◽  
Ting Chen Jiang ◽  
Ming Lian Jiao
Keyword(s):  
Remote Sensing ◽  
Laser Scanning ◽  
Three Dimensional ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Filtering Algorithms ◽  
Filtering Algorithm ◽  
Digital Elevation ◽  
City Model ◽  
Elevation Model

Airborne LiDAR is a new kind of surveying technology of remote sensing which developed rapidly during recent years. Raw laser scanning point clouds data include terrain points, building points, vegetation points, outlier points, etc.. In order to generate digital elevation model (DEM) and three-dimensional city model,these point clouds data must be filtered. Mathematical morphology based filtering algorithm, slope based filtering algorithm, TIN based filtering algorithm, moving surface based filtering algorithm, scanning lines based filtering algorithm and so on several representative filtering algorithms for LiDAR point clouds data have been introduced and discussed and contrasted in this paper. Based on these algorithms summarize the studying progresss about the filtering algorithm of airborne LiDAR point clouds data in home and abroad. In the end, the paper gives an expectation which will provides a reference for the following relative study.

scholarly journals Performance Comparison of Filtering Algorithms for High-Density Airborne LiDAR Point Clouds over Complex LandScapes

2021 ◽  
Vol 13 (14) ◽  
pp. 2663
Author(s):  
Chuanfa Chen ◽  
Jiaojiao Guo ◽  
Huiming Wu ◽  
Yanyan Li ◽  
Bo Shi
Keyword(s):  
Urban Areas ◽  
Total Error ◽  
Point Clouds ◽  
Performance Comparison ◽  
High Density ◽  
Airborne Lidar ◽  
Type I ◽  
Filtering Algorithms ◽  
Study Sites ◽  
Complex Landscapes

Airborne light detection and ranging (LiDAR) technology has become the mainstream data source in geosciences and environmental sciences. Point cloud filtering is a prerequisite for almost all LiDAR-based applications. However, it is challenging to select a suitable filtering algorithm for handling high-density point clouds over complex landscapes. Therefore, to determine an appropriate filter on a specific environment, this paper comparatively assessed the performance of five representative filtering algorithms on six study sites with different terrain characteristics, where three plots are located in urban areas and three in forest areas. The representative filtering methods include simple morphological filter (SMRF), multiresolution hierarchical filter (MHF), slope-based filter (SBF), progressive TIN densification (PTD) and segmentation-based filter (SegBF). Results demonstrate that SMRF performs the best in urban areas, and compared to MHF, SBF, PTD and SegBF, the total error of SMRF is reduced by 1.38%, 48.21%, 48.25% and 31.03%, respectively. MHF outperforms the others in forest areas, and compared to SMRF, SBF, PTD and SegBF, the total error of MHF is reduced by 1.98%, 35.87%, 45.11% and 9.42%, respectively. Moreover, both SMRF and MHF keep a good balance between type I and II errors, which makes the produced DEMs much similar to the references. Overall, SMRF and MHF are recommended for urban and forest areas, respectively, and MHF averagely performs slightly better than SMRF on all areas with respect to kappa coefficient.

Gaussian Mixture Model of Ground Filtering Based on Hierarchical Curvature Constraints for Airborne Lidar Point Clouds

2021 ◽  
Vol 87 (9) ◽  
pp. 615-630
Author(s):  
Longjie Ye ◽  
Ka Zhang ◽  
Wen Xiao ◽  
Yehua Sheng ◽  
Dong Su ◽  
...  
Keyword(s):  
Gaussian Mixture Model ◽  
Mixture Model ◽  
Expectation Maximization ◽  
Latent Variables ◽  
Expectation Maximization Algorithm ◽  
Point Clouds ◽  
Gaussian Mixture ◽  
Airborne Lidar ◽  
Height Difference ◽  
Hierarchical Constraints

This paper proposes a Gaussian mixture model of a ground filtering method based on hierarchical curvature constraints. Firstly, the thin plate spline function is iteratively applied to interpolate the reference surface. Secondly, gradually changing grid size and curvature threshold are used to construct hierarchical constraints. Finally, an adaptive height difference classifier based on the Gaussian mixture model is proposed. Using the latent variables obtained by the expectation-maximization algorithm, the posterior probability of each point is computed. As a result, ground and objects can be marked separately according to the calculated possibility. 15 data samples provided by the International Society for Photogrammetry and Remote Sensing are used to verify the proposed method, which is also compared with eight classical filtering algorithms. Experimental results demonstrate that the average total errors and average Cohen's kappa coefficient of the proposed method are 6.91% and 80.9%, respectively. In general, it has better performance in areas with terrain discontinuities and bridges.

scholarly journals Hybrid Overlap Filter for LiDAR Point Clouds Using Free Software

2020 ◽  
Vol 12 (7) ◽  
pp. 1051 ◽  
Author(s):  
Sandra Buján ◽  
Miguel Cordero ◽  
David Miranda
Keyword(s):  
Total Error ◽  
Digital Terrain Model ◽  
Point Clouds ◽  
Flow Diagram ◽  
Software Environment ◽  
Filtering Algorithms ◽  
Filtering Method ◽  
Hybrid Filtering ◽  
Seed Points ◽  
Reference Samples

Despite the large amounts of resources destined to developing filtering algorithms of LiDAR point clouds in order to obtain a Digital Terrain Model (DTM), the task remains a challenge. As a society advancing towards the democratization of information and collaborative processes, the researchers should not only focus on improving the efficacy of filters, but should also consider the users’ needs with a view toward improving the usability and accessibility of the filters in order to develop tools that will provide solutions to the challenges facing this field of study. In this work, we describe the Hybrid Overlap Filter (HyOF), a new filtering algorithm implemented in the free R software environment. The flow diagram of HyOF differs in the following ways from that of other filters developed to date: (1) the algorithm is formed by a combination of sequentially operating functions (i.e., the output of the first function provides the input of the second), which are capable of functioning independently and thus enabling integration of these functions with other filtering algorithms; (2) the variable penetrability is defined and used, along with slope and elevation, to identify ground points; (3) prior to selection of the seed points, the original point cloud is processed with the aim of removing points corresponding to buildings; and (4) a new method based on a moving window, with longitudinal overlap between windows and transverse overlap between passes, is used to select the seed points. Our hybrid filtering method is tested using 15 reference samples acquired by the International Society of Photogrammetry and Remote Sensing (ISPRS) and is evaluated in comparison with 33 existing filtering algorithms. The results show that our hybrid filtering method produces an average total error of 3.34% and an average Kappa coefficient of 92.62%. The proposed algorithm is one of the most accurate filters that has been tested with the ISPRS reference samples.

scholarly journals Investigating Performance Of Airborne Lidar Data Filtering With Triangular Irregular Network (TIN) Algorithm

2014 ◽  
Vol XL-7 ◽  
pp. 199-202 ◽  
Author(s):  
M. Uysal ◽  
N. Polat
Keyword(s):  
Urban Area ◽  
Rural Area ◽  
Land Surface ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Lidar System ◽  
Filtering Algorithm ◽  
Topographic Features ◽  
Comparison Results ◽  
Triangular Irregular Network

Digital Elevation Model (DEM) is an important topographic product and essential demand for many applications. Traditional methods for creating DEM are very costly and time consuming because of land surveying. In time, Photogrammetry has become one of the major methods to generate DEM. Recently, airborne Light Detection and Ranging (LIDAR) system has become a powerful way to produce a DEM due to advantage of collecting three-dimensional information very effectively over a large area by means of precision and time. <br><br> Airborne LIDAR system collects information not only from land surface but also from every object between plane and terrain that can reflect the laser beam. So filtering out nonground points from raw point clouds is the major step of DEM generation. There are many filtering algorithm due to several factors that affect the filtering prosedures. The performanses of these filters change based on the topographic features of area.One of these algorithm is called Triangular Irregular Network (TIN). <br><br> In this study the TIN algorithm is used to filter Lidar point cloud that are collected from two different sites. While one of these sites is a rural area, the other site is an urban area; therefore these sites have different topographic features. In addition, the reference DEMs are available for these sites. In order to test the performance of TIN algorithm, the Lidar point clouds are filtered and used to generate DEM for the sites. Finally, the generated DEM are compared with the reference DEM for each site. The comparison results show that the TIN filtering algorithm perform more effectively in urban area than rural area in terms of correlations with reference DEMs.

scholarly journals Fast Ground Filtering of Airborne LiDAR Data Based on Iterative Scan-Line Spline Interpolation

2019 ◽  
Vol 11 (19) ◽  
pp. 2256 ◽  
Author(s):  
Jorge Martínez Sánchez ◽  
Álvaro Váquez Álvarez ◽  
David López Vilariño ◽  
Francisco Fernández Rivera ◽  
José Carlos Cabaleiro Domínguez ◽  
...  
Keyword(s):  
Real Time ◽  
Spline Interpolation ◽  
Total Error ◽  
Hardware Acceleration ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Lidar Data ◽  
Wide Range ◽  
Scan Line ◽  
Airborne Lidar Data

Over the last two decades, a wide range of applications have been developed from Light Detection and Ranging (LiDAR) point clouds. Most LiDAR-derived products require the distinction between ground and non-ground points. Because of this, ground filtering its being one of the most studied topics in the literature and robust methods are nowadays available. However, these methods have been designed to work with offline data and they are generally not well suited for real-time scenarios. Aiming to address this issue, this paper proposes an efficient method for ground filtering of airborne LiDAR data based on scan-line processing. In our proposal, an iterative 1-D spline interpolation is performed in each scan line sequentially. The final spline knots of a scan line are taken into account for the next scan line, so that valuable 2-D information is also considered without compromising computational efficiency. Points are labelled into ground and non-ground by analysing their residuals to the final spline. When tested against synthetic ground truth, the method yields a mean kappa value of 88.59% and a mean total error of 0.50%. Experiments with real data also show satisfactory results under visual inspection. Performance tests on a workstation show that the method can process up to 1 million points per second. The original implementation was ported into a low-cost development board to demonstrate its feasibility to run in embedded systems, where throughput was improved by using programmable logic hardware acceleration. Analysis shows that real-time filtering is possible in a high-end board prototype, as it can process the amount of points per second that current lightweight scanners acquire with low-energy consumption.

scholarly journals High-Resolution Terrain Modeling Using Airborne LiDAR Data with Transfer Learning

2021 ◽  
Vol 13 (17) ◽  
pp. 3448
Author(s):  
Huxiong Li ◽  
Weiya Ye ◽  
Jun Liu ◽  
Weikai Tan ◽  
Saied Pirasteh ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Total Error ◽  
Digital Terrain Model ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Type I ◽  
Type Ii ◽  
Terrain Model ◽  
Digital Terrain ◽  
Airborne Lidar Data

This study presents a novel workflow for automated Digital Terrain Model (DTM) extraction from Airborne LiDAR point clouds based on a convolutional neural network (CNN), considering a transfer learning approach. The workflow consists of three parts: feature image generation, transfer learning using ResNet, and interpolation. First, each point is transformed into a featured image based on its elevation differences with neighboring points. Then, the feature images are classified into ground and non-ground using ImageNet pretrained ResNet models. The ground points are extracted by remapping each feature image to its corresponding points. Last, the extracted ground points are interpolated to generate a continuous elevation surface. We compared the proposed workflow with two traditional filters, namely the Progressive Morphological Filter (PMF) and the Progressive Triangulated Irregular Network Densification (PTD). Our results show that the proposed workflow establishes an advantageous DTM extraction accuracy with yields of only 0.52%, 4.84%, and 2.43% for Type I, Type II, and the total error, respectively. In comparison, Type I, Type II, and the total error for PMF are 7.82%, 11.60%, and 9.48% and for PTD 1.55%, 5.37%, and 3.22%, respectively. The root means square error (RMSE) for the 1 m resolution interpolated DTM is only 7.3 cm. Moreover, we conducted a qualitative analysis to investigate the reliability and limitations of the proposed workflow.

scholarly journals Terrestrial Laser Scanning for Vegetation Analyses with a Special Focus on Savannas

2021 ◽  
Vol 13 (3) ◽  
pp. 507
Author(s):  
Tasiyiwa Priscilla Muumbe ◽  
Jussi Baade ◽  
Jenia Singh ◽  
Christiane Schmullius ◽  
Christian Thau
Keyword(s):  
Remote Sensing ◽  
Vegetation Structure ◽  
Laser Scanning ◽  
Spatial Scales ◽  
Terrestrial Laser Scanning ◽  
Point Clouds ◽  
Parameter Extraction ◽  
Airborne Lidar ◽  
Special Focus ◽  
Vegetation Parameter

Savannas are heterogeneous ecosystems, composed of varied spatial combinations and proportions of woody and herbaceous vegetation. Most field-based inventory and remote sensing methods fail to account for the lower stratum vegetation (i.e., shrubs and grasses), and are thus underrepresenting the carbon storage potential of savanna ecosystems. For detailed analyses at the local scale, Terrestrial Laser Scanning (TLS) has proven to be a promising remote sensing technology over the past decade. Accordingly, several review articles already exist on the use of TLS for characterizing 3D vegetation structure. However, a gap exists on the spatial concentrations of TLS studies according to biome for accurate vegetation structure estimation. A comprehensive review was conducted through a meta-analysis of 113 relevant research articles using 18 attributes. The review covered a range of aspects, including the global distribution of TLS studies, parameters retrieved from TLS point clouds and retrieval methods. The review also examined the relationship between the TLS retrieval method and the overall accuracy in parameter extraction. To date, TLS has mainly been used to characterize vegetation in temperate, boreal/taiga and tropical forests, with only little emphasis on savannas. TLS studies in the savanna focused on the extraction of very few vegetation parameters (e.g., DBH and height) and did not consider the shrub contribution to the overall Above Ground Biomass (AGB). Future work should therefore focus on developing new and adjusting existing algorithms for vegetation parameter extraction in the savanna biome, improving predictive AGB models through 3D reconstructions of savanna trees and shrubs as well as quantifying AGB change through the application of multi-temporal TLS. The integration of data from various sources and platforms e.g., TLS with airborne LiDAR is recommended for improved vegetation parameter extraction (including AGB) at larger spatial scales. The review highlights the huge potential of TLS for accurate savanna vegetation extraction by discussing TLS opportunities, challenges and potential future research in the savanna biome.

scholarly journals Cloth simulation-based construction of pit-free canopy height models from airborne LiDAR data

2020 ◽  
Vol 7 (1) ◽  
Author(s):  
Wuming Zhang ◽  
Shangshu Cai ◽  
Xinlian Liang ◽  
Jie Shao ◽  
Ronghai Hu ◽  
...  
Keyword(s):  
Tree Height ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Canopy Height ◽  
Lidar Data ◽  
Cloth Simulation ◽  
Simulation Based ◽  
Maximum Tree Height ◽  
Airborne Lidar Data ◽  
Better Than

Abstract Background The universal occurrence of randomly distributed dark holes (i.e., data pits appearing within the tree crown) in LiDAR-derived canopy height models (CHMs) negatively affects the accuracy of extracted forest inventory parameters. Methods We develop an algorithm based on cloth simulation for constructing a pit-free CHM. Results The proposed algorithm effectively fills data pits of various sizes whilst preserving canopy details. Our pit-free CHMs derived from point clouds at different proportions of data pits are remarkably better than those constructed using other algorithms, as evidenced by the lowest average root mean square error (0.4981 m) between the reference CHMs and the constructed pit-free CHMs. Moreover, our pit-free CHMs show the best performance overall in terms of maximum tree height estimation (average bias = 0.9674 m). Conclusion The proposed algorithm can be adopted when working with different quality LiDAR data and shows high potential in forestry applications.

Sign in / Sign up

Export Citation Format

Share Document