scholarly journals A Parametric Space Approach to the Computation of Multi-scale Geometric Features

2015 ◽  
Keyword(s):  
Geometric Features ◽  
Parametric Space ◽  
Multi Scale ◽  
Space Approach

scholarly journals Separating Leaf and Wood Points in Terrestrial Laser Scanning Data Using Multiple Optimal Scales

Sensors ◽  
2019 ◽  
Vol 19 (8) ◽  
pp. 1852 ◽  
Author(s):  
Junjie Zhou ◽  
Hongqiang Wei ◽  
Guiyun Zhou ◽  
Lihui Song
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Terrestrial Laser Scanning ◽  
Geometric Features ◽  
Optimization Strategy ◽  
Running Time ◽  
Multi Scale ◽  
Scale Method ◽  
Optimal Scale

The separation of leaf and wood points is an essential preprocessing step for extracting many of the parameters of a tree from terrestrial laser scanning data. The multi-scale method and the optimal scale method are two of the most widely used separation methods. In this study, we extend the optimal scale method to the multi-optimal-scale method, adaptively selecting multiple optimal scales for each point in the tree point cloud to increase the distinctiveness of extracted geometric features. Compared with the optimal scale method, our method achieves higher separation accuracy. Compared with the multi-scale method, our method achieves more stable separation accuracy with a limited number of optimal scales. The running time of our method is greatly reduced when the optimization strategy is applied.

scholarly journals CLASSIFICATION OF AIRBORNE LASER SCANNING DATA USING GEOMETRIC MULTI-SCALE FEATURES AND DIFFERENT NEIGHBOURHOOD TYPES

2016 ◽  
Vol III-3 ◽  
pp. 169-176 ◽  
Author(s):  
R. Blomley ◽  
B. Jutzi ◽  
M. Weinmann
Keyword(s):  
Feature Extraction ◽  
Laser Scanning ◽  
Benchmark Dataset ◽  
Airborne Laser Scanning ◽  
Geometric Features ◽  
Multi Scale ◽  
Airborne Laser ◽  
Single Scale ◽  
Detailed Evaluation

In this paper, we address the classification of airborne laser scanning data. We present a novel methodology relying on the use of complementary types of geometric features extracted from multiple local neighbourhoods of different scale and type. To demonstrate the performance of our methodology, we present results of a detailed evaluation on a standard benchmark dataset and we show that the consideration of multi-scale, multi-type neighbourhoods as the basis for feature extraction leads to improved classification results in comparison to single-scale neighbourhoods as well as in comparison to multi-scale neighbourhoods of the same type.

scholarly journals SEMANTIC SEGMENTATION OF AERIAL IMAGERY VIA MULTI-SCALE SHUFFLING CONVOLUTIONAL NEURAL NETWORKS WITH DEEP SUPERVISION

2018 ◽  
Vol IV-1 ◽  
pp. 29-36 ◽  
Author(s):  
K. Chen ◽  
M. Weinmann ◽  
X. Sun ◽  
M. Yan ◽  
S. Hinz ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Semantic Segmentation ◽  
Aerial Imagery ◽  
Geometric Features ◽  
Feature Maps ◽  
Multi Scale ◽  
Intermediate Layers ◽  
Segmentation Task ◽  
The Impact

<p><strong>Abstract.</strong> In this paper, we address the semantic segmentation of aerial imagery based on the use of multi-modal data given in the form of true orthophotos and the corresponding Digital Surface Models (DSMs). We present the Deeply-supervised Shuffling Convolutional Neural Network (DSCNN) representing a multi-scale extension of the Shuffling Convolutional Neural Network (SCNN) with deep supervision. Thereby, we take the advantage of the SCNN involving the shuffling operator to effectively upsample feature maps and then fuse multiscale features derived from the intermediate layers of the SCNN, which results in the Multi-scale Shuffling Convolutional Neural Network (MSCNN). Based on the MSCNN, we derive the DSCNN by introducing additional losses into the intermediate layers of the MSCNN. In addition, we investigate the impact of using different sets of hand-crafted radiometric and geometric features derived from the true orthophotos and the DSMs on the semantic segmentation task. For performance evaluation, we use a commonly used benchmark dataset. The achieved results reveal that both multi-scale fusion and deep supervision contribute to an improvement in performance. Furthermore, the use of a diversity of hand-crafted radiometric and geometric features as input for the DSCNN does not provide the best numerical results, but smoother and improved detections for several objects.</p>

scholarly journals CLASSIFICATION OF AIRBORNE LASER SCANNING DATA USING GEOMETRIC MULTI-SCALE FEATURES AND DIFFERENT NEIGHBOURHOOD TYPES

2016 ◽  
Vol III-3 ◽  
pp. 169-176
Author(s):  
R. Blomley ◽  
B. Jutzi ◽  
M. Weinmann
Keyword(s):  
Feature Extraction ◽  
Laser Scanning ◽  
Benchmark Dataset ◽  
Airborne Laser Scanning ◽  
Geometric Features ◽  
Multi Scale ◽  
Airborne Laser ◽  
Single Scale ◽  
Detailed Evaluation

In this paper, we address the classification of airborne laser scanning data. We present a novel methodology relying on the use of complementary types of geometric features extracted from multiple local neighbourhoods of different scale and type. To demonstrate the performance of our methodology, we present results of a detailed evaluation on a standard benchmark dataset and we show that the consideration of multi-scale, multi-type neighbourhoods as the basis for feature extraction leads to improved classification results in comparison to single-scale neighbourhoods as well as in comparison to multi-scale neighbourhoods of the same type.

Geometric Features Applied to Composite Surfaces Using Spherical Reparametrization

1999 ◽  
Author(s):  
Paul J. Stewart ◽  
Yifan Chen
Keyword(s):  
Projection Method ◽  
Dimensional Space ◽  
Geometric Features ◽  
Two Dimensional ◽  
Parametric Surface ◽  
Parametric Space ◽  
Curvature Variation ◽  
The Common ◽  
Spherical Projection ◽  
Global Surface

Abstract The Direct Surface Manipulation (DSM) method allows creation and modification of user-specified features on an existing parametric surface. A global surface reparametrization scheme is presented that extends DSM to include features spanning multiple surfaces. This technique globally reparametrizes multiple surfaces onto a two-dimensional space, and the result is used as a global, uniform parametric space to connect the topologically-disconnected surfaces. Features defined on the common parametric space then span multiple surfaces. Spherical projection is presented as an effective projection method to use for the global reparametrization because of its ability to reduce shape distortions for features placed on surfaces with substantial curvature variation.

Computer Assisted Image Reconstruction of Oligomer Structure

1976 ◽  
Vol 34 ◽  
pp. 472-473
Author(s):  
A.M. Jones ◽  
A. Max Fiskin
Keyword(s):  
Three Dimensional ◽  
Depth Of Field ◽  
Computer Assisted ◽  
Projection Data ◽  
Two Dimensional ◽  
Single Particles ◽  
Dimensional Reconstruction ◽  
Computer Assisted Image ◽  
The Fourier Transform ◽  
Space Approach

If the tilt of a specimen can be varied either by the strategy of observing identical particles orientated randomly or by use of a eucentric goniometer stage, three dimensional reconstruction procedures are available (l). If the specimens, such as small protein aggregates, lack periodicity, direct space methods compete favorably in ease of implementation with reconstruction by the Fourier (transform) space approach (2). Regardless of method, reconstruction is possible because useful specimen thicknesses are always much less than the depth of field in an electron microscope. Thus electron images record the amount of stain in columns of the object normal to the recording plates. For single particles, practical considerations dictate that the specimen be tilted precisely about a single axis. In so doing a reconstructed image is achieved serially from two-dimensional sections which in turn are generated by a series of back-to-front lines of projection data.

Sign in / Sign up

Export Citation Format

Share Document