scholarly journals An MFF-SLIC Hybrid Superpixel Segmentation Method with Multi-Source RS Data for Rock Surface Extraction

2019 ◽  
Vol 9 (5) ◽  
pp. 906 ◽  
Author(s):  
Xuefeng Yi ◽  
Rongchun Zhang ◽  
Hao Li ◽  
Yuanyuan Chen
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Distance Measure ◽  
Digital Images ◽  
Measurement Techniques ◽  
Rock Surface ◽  
Segmentation Method ◽  
Surface Extraction ◽  
Superpixel Segmentation ◽  
Relationship Construction

Multi-Source RS data integration is a crucial technology for rock surface extraction in geology. Both Terrestrial laser scanning (TLS) and Photogrammetry are primary non-contact active measurement techniques. In order to extract comprehensive and accurate rock surface information by the integration of TLS point cloud and digital images, the segmentation based on the integrated results generated by registration is the crux. This paper presents a Multi-Features Fusion for Simple Linear Iterative Clustering (MFF-SLIC) hybrid superpixel segmentation algorithm to extract the rock surface accurately. The MFF-SLIC algorithm mainly includes three contents: (1) Mapping relationship construction for TLS point cloud and digital images; (2) Distance measure model establishment with multi-features for initial superpixel segmentation; (3) Hierarchical and optimized clustering for superpixels. The proposed method was verified with the columnar basalt data, which is acquired in Guabushan Geopark in China. The results demonstrate that the segmentation method could be used for rock surface extraction with high precision and efficiency, the result of which would be prepared for further geological statistics and analysis.

Assessment of automatic segmentation accuracy with various point cloud density

2020 ◽  
Vol 961 (7) ◽  
pp. 47-55
Author(s):  
A.G. Yunusov ◽  
A.J. Jdeed ◽  
N.S. Begliarov ◽  
M.A. Elshewy
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Reference Data ◽  
Automatic Segmentation ◽  
Geometric Accuracy ◽  
Segmentation Method ◽  
Segmentation Methods ◽  
Segmentation Accuracy ◽  
Number Of Segments ◽  
Cloud Density

Laser scanning is considered as one of the most useful and fast technologies for modelling. On the other hand, the size of scan results can vary from hundreds to several million points. As a result, the large volume of the obtained clouds leads to complication at processing the results and increases the time costs. One way to reduce the volume of a point cloud is segmentation, which reduces the amount of data from several million points to a limited number of segments. In this article, we evaluated effect on the performance, the accuracy of various segmentation methods and the geometric accuracy of the obtained models at density changes taking into account the processing time. The results of our experiment were compared with reference data in a form of comparative analysis. As a conclusion, some recommendations for choosing the best segmentation method were proposed.

scholarly journals Superpixel Segmentation Based on Anisotropic Edge Strength

2019 ◽  
Vol 5 (6) ◽  
pp. 57 ◽  
Author(s):  
Gang Wang ◽  
Bernard De Baets
Keyword(s):  
Distance Measure ◽  
Saliency Detection ◽  
Experimental Results ◽  
Segmentation Method ◽  
Superpixel Segmentation ◽  
Gaussian Kernels ◽  
First Derivative ◽  
Edge Strength

Superpixel segmentation can benefit from the use of an appropriate method to measure edge strength. In this paper, we present such a method based on the first derivative of anisotropic Gaussian kernels. The kernels can capture the position, direction, prominence, and scale of the edge to be detected. We incorporate the anisotropic edge strength into the distance measure between neighboring superpixels, thereby improving the performance of an existing graph-based superpixel segmentation method. Experimental results validate the superiority of our method in generating superpixels over the competing methods. It is also illustrated that the proposed superpixel segmentation method can facilitate subsequent saliency detection.

scholarly journals An Automatic Density Clustering Segmentation Method for Laser Scanning Point Cloud Data of Buildings

2019 ◽  
Vol 2019 ◽  
pp. 1-13
Author(s):  
Jianghong Zhao ◽  
Yan Dong ◽  
Siyu Ma ◽  
Huajun Liu ◽  
Shuangfeng Wei ◽  
...  
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Three Dimensional ◽  
Point Cloud Data ◽  
Segmentation Method ◽  
Cloud Data ◽  
Dbscan Algorithm ◽  
Density Clustering ◽  
Low Efficiency ◽  
Definition Of

Segmentation is an important step in point cloud data feature extraction and three-dimensional modelling. Currently, it is also a challenging problem in point cloud processing. There are some disadvantages of the DBSCAN method, such as requiring the manual definition of parameters and low efficiency when it is used for large amounts of calculation. This paper proposes the AQ-DBSCAN algorithm, which is a density clustering segmentation method combined with Gaussian mapping. The algorithm improves upon the DBSCAN algorithm by solving the problem of automatic estimation of the parameter neighborhood radius. The improved algorithm can carry out density clustering processing quickly by reducing the amount of computation required.

scholarly journals MLS POINT CLOUD SEGMENTATION BASED ON FEATURE POINTS OF SCANLINES

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 1007-1013
Author(s):  
R. Honma ◽  
H. Date ◽  
S. Kanai
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Point Clouds ◽  
Spatial Distance ◽  
Feature Points ◽  
Segmentation Method ◽  
Point Reduction ◽  
Point Cloud Segmentation ◽  
Edge Points ◽  
The Difference

<p><strong>Abstract.</strong> Point clouds acquired using Mobile Laser Scanning (MLS) are applied to extract road information such as curb stones, road markings, and road side objects. In this paper, we present a scanline-based MLS point cloud segmentation method for various road and road side objects. First, end points of the scanline, jump edge points, and corner points are extracted as feature points. The feature points are then interpolated to accurately extract irregular parts consisting of irregularly distributed points such as vegetation. Next, using a point reduction method, additional feature points on a smooth surface are extracted for segmentation at the edges of the curb cut. Finally, points between the feature points are extracted as flat segments on the scanline, and continuing feature points are extracted as irregular segments on the scanline. Furthermore, these segments on the scanline are integrated as flat or irregular regions. In the extraction of the feature points, neighboring points based on the spatial distance are used to avoid being influenced by the difference in the point density. Based on experiments, the effectiveness of the proposed method was indicated based on an application to an MLS point cloud.</p>

scholarly journals LONG LINE CLIFF TOPOGRAPHY MEASUREMENT BY USING STRUCTURE FROM MOTION ALGORITHM IN PHOTOGRAMMETRY

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 5-12
Author(s):  
C. Altuntas
Keyword(s):  
Structure From Motion ◽  
Point Cloud ◽  
Laser Scanning ◽  
3D Model ◽  
Measurement Techniques ◽  
High Accuracy ◽  
Aerial Laser Scanning ◽  
The Earth ◽  
Long Line ◽  
Dense Point

<p><strong>Abstract.</strong> The topography of cliffs and steep slopes must be measured to acquire additional information for landscaping, visualizing changes and taking precautions against natural hazards. The Earth topography has been measured predominantly with photogrammetry, terrestrial/aerial laser scanning or other traditional measurement techniques. The stereo photogrammetry necessitates greater effort to obtain a three-dimensional (3D) model of the imaged surface. Meanwhile, terrestrial or aerial laser scanning can collect high-density measurements of spatial data in a short time. However, the costs of implementing laser scanning instruments are very high. Furthermore, conventional measurement techniques that use total stations require immense effort to collect complete 3D measurements of cliffs. On the other hand, dense image based point cloud using multi-view photogrammetry based on structure from motion (SfM) algorithm is much more effective than the others for measuring the Earth topography. In this study, the cliff topography of an old quarry located in the state of Selcuklu of Konya Province in Turkey was measured by multi-view photogrammetry. The cliff has a continuous length of approximately 600 metres and a height of 25 metres in some places. The 3D model of the cliff was generated with the image based dense point cloud of multi-view photogrammetry. Then 3D dense point cloud model was registered into a local georeference system by using control points (CPs). Because of the long line measurement area, number and localization of the CPs is very important for achieving a high-accuracy to registration into georeferenced system. The registration accuracies were evaluated for different number and distribution of the CPs with the residuals on the check points (ChPs). The high accuracy registration was acquired with uniform distributed 3 and 8 CPs as the residuals of 24.08&amp;thinsp;cm and 23.03&amp;thinsp;cm on the ChPs respectively. The results indicated that 3D measurement of long line cliffs can be performed using multi-view photogrammetry, and the registration should be made with the uniform distributed CPs. In addition, a texture-mapped 3D model and orthophoto images of the cliff surfaces were created for detailed visualization.</p>

scholarly journals SCAN PROFILES BASED METHOD FOR SEGMENTATION AND EXTRACTION OF PLANAR OBJECTS IN MOBILE LASER SCANNING POINT CLOUDS

2016 ◽  
Vol XLI-B3 ◽  
pp. 351-358 ◽  
Author(s):  
Hoang Long Nguyen ◽  
David Belton ◽  
Petra Helmholz
Keyword(s):  
Spatial Data ◽  
Point Cloud ◽  
Laser Scanning ◽  
Laser Scanner ◽  
Point Clouds ◽  
Point Cloud Data ◽  
Segmentation Method ◽  
Cloud Data ◽  
Processing Steps ◽  
Carrier Vehicle

The demand for accurate spatial data has been increasing rapidly in recent years. Mobile laser scanning (MLS) systems have become a mainstream technology for measuring 3D spatial data. In a MLS point cloud, the point clouds densities of captured point clouds of interest features can vary: they can be sparse and heterogeneous or they can be dense. This is caused by several factors such as the speed of the carrier vehicle and the specifications of the laser scanner(s). The MLS point cloud data needs to be processed to get meaningful information e.g. segmentation can be used to find meaningful features (planes, corners etc.) that can be used as the inputs for many processing steps (e.g. registration, modelling) that are more difficult when just using the point cloud. Planar features are dominating in manmade environments and they are widely used in point clouds registration and calibration processes. There are several approaches for segmentation and extraction of planar objects available, however the proposed methods do not focus on properly segment MLS point clouds automatically considering the different point densities. This research presents the extension of the segmentation method based on planarity of the features. This proposed method was verified using both simulated and real MLS point cloud datasets. The results show that planar objects in MLS point clouds can be properly segmented and extracted by the proposed segmentation method.

scholarly journals 3D GEOLOGICAL OUTCROP CHARACTERIZATION: AUTOMATIC DETECTION OF 3D PLANES (AZIMUTH AND DIP) USING LiDAR POINT CLOUDS

2016 ◽  
Vol III-5 ◽  
pp. 105-112 ◽  
Author(s):  
K. Anders ◽  
M. Hämmerle ◽  
G. Miernik ◽  
T. Drews ◽  
A. Escalona ◽  
...  
Keyword(s):  
Spatial Resolution ◽  
Point Cloud ◽  
Laser Scanning ◽  
Spatial Information ◽  
Point Clouds ◽  
Rock Surface ◽  
Cloud Data ◽  
Surface Data ◽  
Data Volume

Terrestrial laser scanning constitutes a powerful method in spatial information data acquisition and allows for geological outcrops to be captured with high resolution and accuracy. A crucial aspect for numerous geologic applications is the extraction of rock surface orientations from the data. This paper focuses on the detection of planes in rock surface data by applying a segmentation algorithm directly to a 3D point cloud. Its performance is assessed considering (1) reduced spatial resolution of data and (2) smoothing in the course of data pre-processing. The methodology is tested on simulations of progressively reduced spatial resolution defined by varying point cloud density. Smoothing of the point cloud data is implemented by modifying the neighborhood criteria during normals estima-tion. The considerable alteration of resulting planes emphasizes the influence of smoothing on the plane detection prior to the actual segmentation. Therefore, the parameter needs to be set in accordance with individual purposes and respective scales of studies. Fur-thermore, it is concluded that the quality of segmentation results does not decline even when the data volume is significantly reduced down to 10%. The azimuth and dip values of individual segments are determined for planes fit to the points belonging to one segment. Based on these results, azimuth and dip as well as strike character of the surface planes in the outcrop are assessed. Thereby, this paper contributes to a fully automatic and straightforward workflow for a comprehensive geometric description of outcrops in 3D.

scholarly journals JOINT CLASSIFICATION OF ALS AND DIM POINT CLOUDS

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 1113-1120
Author(s):  
F. Politz ◽  
M. Sester
Keyword(s):  
Transfer Learning ◽  
Point Cloud ◽  
Laser Scanning ◽  
Measurement Techniques ◽  
Point Clouds ◽  
Aerial Images ◽  
Individual Measurement ◽  
Cloud Data ◽  
Practical Applications

<p><strong>Abstract.</strong> National mapping agencies (NMAs) have to acquire nation-wide Digital Terrain Models on a regular basis as part of their obligations to provide up-to-date data. Point clouds from Airborne Laser Scanning (ALS) are an important data source for this task; recently, NMAs also started deriving Dense Image Matching (DIM) point clouds from aerial images. As a result, NMAs have both point cloud data sources available, which they can exploit for their purposes. In this study, we investigate the potential of transfer learning from ALS to DIM data, so the time consuming step of data labelling can be reduced. Due to their specific individual measurement techniques, both point clouds have various distinct properties such as RGB or intensity values, which are often exploited for classification of either ALS or DIM point clouds. However, those features also hinder transfer learning between these two point cloud types, since they do not exist in the other point cloud type. As the mere 3D point is available in both point cloud types, we focus on transfer learning from an ALS to a DIM point cloud using exclusively the point coordinates. We are tackling the issue of different point densities by rasterizing the point cloud into a 2D grid and take important height features as input for classification. We train an encoder-decoder convolutional neural network with labelled ALS data as a baseline and then fine-tune this baseline with an increasing amount of labelled DIM data. We also train the same network exclusively on all available DIM data as reference to compare our results. We show that only 10% of labelled DIM data increase the classification results notably, which is especially relevant for practical applications.</p>

Sign in / Sign up

Export Citation Format

Share Document