scholarly journals QUALITY IMPROVEMENT FOR AIRBORNE LIDAR DATA FILTERING BASED ON DEEP LEARNING METHOD

2020 ◽  
Vol XLIII-B2-2020 ◽  
pp. 355-360
Author(s):  
T. Yotsumata ◽  
M. Sakamoto ◽  
T. Satoh
Keyword(s):  
Deep Learning ◽  
Ground Surface ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Baseline Method ◽  
Classification Result ◽  
Multiple Networks ◽  
The Individual ◽  
Airborne Lidar Data

Abstract. In this paper, we discuss how to improve the quality of classification results when deep learning is applied for the filtering of airborne LiDAR point cloud. We introduce the baseline method which utilizes convolutional neural network (CNN) based on voxelization, and then we propose three methods to improve the quality of classification result. The first method is data pre-processing that aims to exclude data in advance that is clearly not on the ground surface in order to efficiently extract the ground surface data. Data pre-processing can greatly reduce the number of target points and the subsequent processing can be performed efficiently. It also has the effect of preventing noise-like points floating in the air from being misclassified as the ground surface, as compared to the case without pre-processing. The second method is changing the network structure. In recent years, various networks have been proposed for classifying point clouds. In our study, the baseline is using very simple networks. In order to improve the classification result of the baseline method, the layer depth and the range size of convolution are changed, and we investigated about the improvements of the results. The current discussion can be used as a guidance when considering new networks. The third method is the integration of classification results from multiple networks. We integrated individual results from multiple networks with varying layer depths and convolution sizes, starting with the baseline, and investigated whether the results improved. We observed that even if the individual results were similar, the classification results can be improved by integrating the results.

scholarly journals GROUND POINT FILTERING FROM AIRBORNE LIDAR POINT CLOUDS USING DEEP LEARNING: A PRELIMINARY STUDY

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 1559-1565 ◽  
Author(s):  
E. Janssens-Coron ◽  
E. Guilbert
Keyword(s):  
Deep Learning ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Lidar Data ◽  
Learning Approach ◽  
Preliminary Results ◽  
Ground Point ◽  
Preliminary Study ◽  
Airborne Lidar Data ◽  
Made In

<p><strong>Abstract.</strong> Airborne lidar data is commonly used to generate point clouds over large areas. These points can be classified into different categories such as ground, building, vegetation, etc. The first step for this is to separate ground points from non-ground points. Existing methods rely mainly on TIN densification but there performance varies with the type of terrain and relies on the user’s experience who adjusts parameters accordingly. An alternative may be on the use of a deep learning approach that would limit user’s intervention. Hence, in this paper, we assess a deep learning architecture, PointNet, that applies directly to point clouds. Our preliminary results show mitigating classification rates and further investigation is required to properly train the system and improve the robustness, showing issues with the choices we made in the preprocessing. Nonetheless, our analysis suggests that it is necessary to enrich the architecture of the network to integrate the notion of neighbourhood at different scales in order to increase the accuracy and the robustness of the treatment as well as its capacity to treat data from different geographical contexts.</p>

Données LiDAR aériennes: pertinence de l'interprétation visuelle pour la foresterie

2017 ◽  
Vol 168 (3) ◽  
pp. 127-133
Author(s):  
Matthew Parkan
Keyword(s):  
Cross Sections ◽  
Forest Edge ◽  
Airborne Lidar ◽  
Visual Interpretation ◽  
Individual Tree ◽  
Small Areas ◽  
Interactive 3D ◽  
The Individual ◽  
Airborne Lidar Data ◽  
Stem Position

Airborne LiDAR data: relevance of visual interpretation for forestry Airborne LiDAR surveys are particularly well adapted to map, study and manage large forest extents. Products derived from this technology are increasingly used by managers to establish a general diagnosis of the condition of forests. Less common is the use of these products to conduct detailed analyses on small areas; for example creating detailed reference maps like inventories or timber marking to support field operations. In this context, the use of direct visual interpretation is interesting, because it is much easier to implement than automatic algorithms and allows a quick and reliable identification of zonal (e.g. forest edge, deciduous/persistent ratio), structural (stratification) and point (e.g. tree/stem position and height) features. This article examines three important points which determine the relevance of visual interpretation: acquisition parameters, interactive representation and identification of forest characteristics. It is shown that the use of thematic color maps within interactive 3D point cloud and/or cross-sections makes it possible to establish (for all strata) detailed and accurate maps of a parcel at the individual tree scale.

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.

scholarly journals Building Extraction from Airborne LiDAR Data Based on Multi-Constraints Graph Segmentation

2021 ◽  
Vol 13 (18) ◽  
pp. 3766
Author(s):  
Zhenyang Hui ◽  
Zhuoxuan Li ◽  
Penggen Cheng ◽  
Yao Yevenyo Ziggah ◽  
JunLin Fan
Keyword(s):  
Point Clouds ◽  
Optimization Method ◽  
Extraction Methods ◽  
Urban Environments ◽  
Airborne Lidar ◽  
Lidar Data ◽  
Building Extraction ◽  
Existing Building ◽  
Significant Step ◽  
Airborne Lidar Data

Building extraction from airborne Light Detection and Ranging (LiDAR) point clouds is a significant step in the process of digital urban construction. Although the existing building extraction methods perform well in simple urban environments, when encountering complicated city environments with irregular building shapes or varying building sizes, these methods cannot achieve satisfactory building extraction results. To address these challenges, a building extraction method from airborne LiDAR data based on multi-constraints graph segmentation was proposed in this paper. The proposed method mainly converted point-based building extraction into object-based building extraction through multi-constraints graph segmentation. The initial extracted building points were derived according to the spatial geometric features of different object primitives. Finally, a multi-scale progressive growth optimization method was proposed to recover some omitted building points and improve the completeness of building extraction. The proposed method was tested and validated using three datasets provided by the International Society for Photogrammetry and Remote Sensing (ISPRS). Experimental results show that the proposed method can achieve the best building extraction results. It was also found that no matter the average quality or the average F1 score, the proposed method outperformed ten other investigated building extraction methods.

scholarly journals A SEGMENT-BASED APPROACH FOR DTM DERIVATION OF AIRBORNE LIDAR DATA

2016 ◽  
Vol XLI-B1 ◽  
pp. 115-120
Author(s):  
Dejin Tang ◽  
Xiaoming Zhou ◽  
Jie Jiang ◽  
Caiping Li
Keyword(s):  
Confidence Level ◽  
Spatial Location ◽  
Point Clouds ◽  
Airborne Lidar ◽  
High Confidence ◽  
Confidence Levels ◽  
Segment Distribution ◽  
Adjacent Segments ◽  
Airborne Lidar Data ◽  
High Confidence Level

With the characteristics of LIDAR system, raw point clouds represent both terrain and non-terrain surface. In order to generate DTM, the paper introduces one improved filtering method based on the segment-based algorithms. The method generates segments by clustering points based on surface fitting and uses topological and geometric properties for classification. In the process, three major steps are involved. First, the whole datasets is split into several small overlapping tiles. For each tile, by removing wall and vegetation points, accurate segments are found. The segments from all tiles are assigned unique segment number. In the following step, topological descriptions for the segment distribution pattern and height jump between adjacent segments are identified in each tile. Based on the topology and geometry, segment-based filtering algorithm is performed for classification in each tile. Then, based on the spatial location of the segment in one tile, two confidence levels are assigned to the classified segments. The segments with low confidence level are because of losing geometric or topological information in one tile. Thus, a combination algorithm is generated to detect corresponding parts of incomplete segment from multiple tiles. Then another classification algorithm is performed for these segments. The result of these segments will have high confidence level. After that, all the segments in one tile have high confidence level of classification result. The final DTM will add all the terrain segments and avoid duplicate points. At the last of the paper, the experiment show the filtering result and be compared with the other classical filtering methods, the analysis proves the method has advantage in the precision of DTM. But because of the complicated algorithms, the processing speed is little slower, that is the future improvement which should been researched.

scholarly journals ACCURACY ASSESSMENT OF LIDAR-DERIVED DIGITAL TERRAIN MODEL (DTM) WITH DIFFERENT SLOPE AND CANOPY COVER IN TROPICAL FOREST REGION

2015 ◽  
Vol II-2/W2 ◽  
pp. 183-189 ◽  
Author(s):  
M. R. M. Salleh ◽  
Z. Ismail ◽  
M. Z. A. Rahman
Keyword(s):  
Tropical Forest ◽  
Digital Terrain Model ◽  
Canopy Cover ◽  
Airborne Lidar ◽  
Lidar Data ◽  
Terrain Model ◽  
Digital Terrain ◽  
Terrain Slope ◽  
Airborne Lidar Data

Airborne Light Detection and Ranging (LiDAR) technology has been widely used recent years especially in generating high accuracy of Digital Terrain Model (DTM). High density and good quality of airborne LiDAR data promises a high quality of DTM. This study focussing on the analysing the error associated with the density of vegetation cover (canopy cover) and terrain slope in a LiDAR derived-DTM value in a tropical forest environment in Bentong, State of Pahang, Malaysia. Airborne LiDAR data were collected can be consider as low density captured by Reigl system mounted on an aircraft. The ground filtering procedure use adaptive triangulation irregular network (ATIN) algorithm technique in producing ground points. Next, the ground control points (GCPs) used in generating the reference DTM and these DTM was used for slope classification and the point clouds belong to non-ground are then used in determining the relative percentage of canopy cover. The results show that terrain slope has high correlation for both study area (0.993 and 0.870) with the RMSE of the LiDAR-derived DTM. This is similar to canopy cover where high value of correlation (0.989 and 0.924) obtained. This indicates that the accuracy of airborne LiDAR-derived DTM is significantly affected by terrain slope and canopy caver of study area.

scholarly journals INTEGRATING SMARTPHONE IMAGES AND AIRBORNE LIDAR DATA FOR COMPLETE URBAN BUILDING MODELLING

2016 ◽  
Vol XLI-B5 ◽  
pp. 741-747 ◽  
Author(s):  
Shenman Zhang ◽  
Jie Shan ◽  
Zhichao Zhang ◽  
Jixing Yan ◽  
Yaolin Hou
Keyword(s):  
Point Cloud ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Image Point ◽  
Lidar Data ◽  
Building Model ◽  
Building Facades ◽  
Single Data ◽  
Airborne Lidar Data ◽  
Building Reconstruction

A complete building model reconstruction needs data collected from both air and ground. The former often has sparse coverage on building façades, while the latter usually is unable to observe the building rooftops. Attempting to solve the missing data issues in building reconstruction from single data source, we describe an approach for complete building reconstruction that integrates airborne LiDAR data and ground smartphone imagery. First, by taking advantages of GPS and digital compass information embedded in the image metadata of smartphones, we are able to find airborne LiDAR point clouds for the corresponding buildings in the images. In the next step, Structure-from-Motion and dense multi-view stereo algorithms are applied to generate building point cloud from multiple ground images. The third step extracts building outlines respectively from the LiDAR point cloud and the ground image point cloud. An automated correspondence between these two sets of building outlines allows us to achieve a precise registration and combination of the two point clouds, which ultimately results in a complete and full resolution building model. The developed approach overcomes the problem of sparse points on building façades in airborne LiDAR and the deficiency of rooftops in ground images such that the merits of both datasets are utilized.

scholarly journals A NOVEL DENOISING ALGORITHM FOR AIRBORNE LIDAR POINT CLOUD BASED ON EMPIRICAL MODE DECOMPOSITION

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 1021-1025
Author(s):  
Z. Hui ◽  
P. Cheng ◽  
L. Wang ◽  
Y. Xia ◽  
H. Hu ◽  
...  
Keyword(s):  
Empirical Mode Decomposition ◽  
Point Cloud ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Morphological Operations ◽  
Intrinsic Mode Functions ◽  
Mode Decomposition ◽  
Processing Step ◽  
Cloud Applications

<p><strong>Abstract.</strong> Denoising is a key pre-processing step for many airborne LiDAR point cloud applications. However, the previous algorithms have a number of problems, which affect the quality of point cloud post-processing, such as DTM generation. In this paper, a novel automated denoising algorithm is proposed based on empirical mode decomposition to remove outliers from airborne LiDAR point cloud. Comparing with traditional point cloud denoising algorithms, the proposed method can detect outliers from a signal processing perspective. Firstly, airborne LiDAR point clouds are decomposed into a series of intrinsic mode functions with the help of morphological operations, which would significantly decrease the computational complexity. By applying OTSU algorithm to these intrinsic mode functions, noise-dominant components can be detected and filtered. Finally, outliers are detected automatically by comparing observed elevations and reconstructed elevations. Three datasets located at three different cities in China were used to verify the validity and robustness of the proposed method. The experimental results demonstrate that the proposed method removes both high and low outliers effectively with various terrain features while preserving useful ground details.</p>

Sign in / Sign up

Export Citation Format

Share Document