scholarly journals Snake-Based Model for Automatic Roof Boundary Extraction in the Object Space Integrating a High-Resolution Aerial Images Stereo Pair and 3D Roof Models

2021 ◽  
Vol 13 (8) ◽  
pp. 1429
Author(s):  
Michelle S. Y. Ywata ◽  
Aluir P. Dal Poz ◽  
Milton H. Shimabukuro ◽  
Henrique C. de Oliveira
Keyword(s):  
High Resolution ◽  
Spatial Information ◽  
Contextual Information ◽  
Three Dimensional ◽  
Extraction Methods ◽  
Aerial Images ◽  
Stereo Pair ◽  
Boundary Extraction ◽  
Building Roof ◽  
Object Space

The accelerated urban development over the last decades has made it necessary to update spatial information rapidly and constantly. Therefore, cities’ three-dimensional models have been widely used as a study base for various urban problems. However, although many efforts have been made to develop new building extraction methods, reliable and automatic extraction is still a major challenge for the remote sensing and computer vision communities, mainly due to the complexity and variability of urban scenes. This paper presents a method to extract building roof boundaries in the object space by integrating a high-resolution aerial images stereo pair, three-dimensional roof models reconstructed from light detection and ranging (LiDAR) data, and contextual information of the scenes involved. The proposed method focuses on overcoming three types of common problems that can disturb the automatic roof extraction in the urban environment: perspective occlusions caused by high buildings, occlusions caused by vegetation covering the roof, and shadows that are adjacent to the roofs, which can be misinterpreted as roof edges. For this, an improved Snake-based mathematical model is developed considering the radiometric and geometric properties of roofs to represent the roof boundary in the image space. A new approach for calculating the corner response and a shadow compensation factor was added to the model. The created model is then adapted to represent the boundaries in the object space considering a stereo pair of aerial images. Finally, the optimal polyline, representing a selected roof boundary, is obtained by optimizing the proposed Snake-based model using a dynamic programming (DP) approach considering the contextual information of the scene. The results showed that the proposed method works properly in boundary extraction of roofs with occlusion and shadows areas, presenting completeness and correctness average values above 90%, RMSE average values below 0.5 m for E and N components, and below 1 m for H component.

scholarly journals Multiscale Semantic Feature Optimization and Fusion Network for Building Extraction Using High-Resolution Aerial Images and LiDAR Data

2021 ◽  
Vol 13 (13) ◽  
pp. 2473
Author(s):  
Qinglie Yuan ◽  
Helmi Zulhaidi Mohd Shafri ◽  
Aidi Hizami Alias ◽  
Shaiful Jahari Hashim
Keyword(s):  
High Resolution ◽  
Large Scale ◽  
Spatial Information ◽  
Feature Fusion ◽  
Aerial Images ◽  
Semantic Gap ◽  
Superior Performance ◽  
Lidar Data ◽  
Building Extraction ◽  
Hierarchical Features

Automatic building extraction has been applied in many domains. It is also a challenging problem because of the complex scenes and multiscale. Deep learning algorithms, especially fully convolutional neural networks (FCNs), have shown robust feature extraction ability than traditional remote sensing data processing methods. However, hierarchical features from encoders with a fixed receptive field perform weak ability to obtain global semantic information. Local features in multiscale subregions cannot construct contextual interdependence and correlation, especially for large-scale building areas, which probably causes fragmentary extraction results due to intra-class feature variability. In addition, low-level features have accurate and fine-grained spatial information for tiny building structures but lack refinement and selection, and the semantic gap of across-level features is not conducive to feature fusion. To address the above problems, this paper proposes an FCN framework based on the residual network and provides the training pattern for multi-modal data combining the advantage of high-resolution aerial images and LiDAR data for building extraction. Two novel modules have been proposed for the optimization and integration of multiscale and across-level features. In particular, a multiscale context optimization module is designed to adaptively generate the feature representations for different subregions and effectively aggregate global context. A semantic guided spatial attention mechanism is introduced to refine shallow features and alleviate the semantic gap. Finally, hierarchical features are fused via the feature pyramid network. Compared with other state-of-the-art methods, experimental results demonstrate superior performance with 93.19 IoU, 97.56 OA on WHU datasets and 94.72 IoU, 97.84 OA on the Boston dataset, which shows that the proposed network can improve accuracy and achieve better performance for building extraction.

scholarly journals Vehicle Spatial Distribution and 3D Trajectory Extraction Algorithm in a Cross-Camera Traffic Scene

Sensors ◽  
2020 ◽  
Vol 20 (22) ◽  
pp. 6517
Author(s):  
Xinyao Tang ◽  
Huansheng Song ◽  
Wei Wang ◽  
Yanni Yang
Keyword(s):  
Spatial Distribution ◽  
Spatial Information ◽  
Three Dimensional ◽  
Physical Space ◽  
Extraction Methods ◽  
Geometric Constraints ◽  
Panoramic Image ◽  
Trajectory Data ◽  
Comparison Algorithms ◽  
Bounding Boxes

The three-dimensional trajectory data of vehicles have important practical meaning for traffic behavior analysis. To solve the problems of narrow visual angle in single-camera scenes and lack of continuous trajectories in 3D space by current cross-camera trajectory extraction methods, we propose an algorithm of vehicle spatial distribution and 3D trajectory extraction in this paper. First, a panoramic image of a road with spatial information is generated based on camera calibration, which is used to convert cross-camera perspectives into 3D physical space. Then, we choose YOLOv4 to obtain 2D bounding boxes of vehicles in cross-camera scenes. Based on the above information, 3D bounding boxes around vehicles are built with geometric constraints which are used to obtain projection centroids of vehicles. Finally, by calculating the spatial distribution of projection centroids in the panoramic image, 3D trajectories of vehicles are extracted. The experimental results indicate that our algorithm can effectively complete vehicle spatial distribution and 3D trajectory extraction in various traffic scenes, which outperforms other comparison algorithms.

A System for Computer-based Reconstruction of 3-Dimensional Structures from Serial Tissue Sections: an Application to the Study of Normal and Neoplastic Mammary Gland Biology

2001 ◽  
Vol 7 (S2) ◽  
pp. 964-965
Author(s):  
Rodrigo Fernandez-Gonzalez ◽  
Arthur Jones ◽  
Enrique Garcia-Rodriguez ◽  
Davis Knowles ◽  
Damir Sudar ◽  
...  
Keyword(s):  
Mammary Gland ◽  
High Resolution ◽  
Imaging System ◽  
Contextual Information ◽  
Three Dimensional ◽  
Tissue Level ◽  
Disease Heterogeneity ◽  
Tissue Sections ◽  
Virtual Reconstruction ◽  
Analytical Microscopy

Tissue heterogeneity and three-dimensionality are generally neglected by most traditional analytical microscopy methods in Biology. These often disregard contextual information important for understanding most biological systems. in breast cancer, which is a tissue level disease, heterogeneity and three dimensionality are at the very base of cancer initiation and clonal progression. Thus, a three dimensional quantitative system that allows low resolution virtual reconstruction of the mammary gland from serial sections, followed by high resolution cell-level reconstruction and quantitative analysis of the ductal epithelium emerges as an essential tool in studying the disease. We present here a distributed microscopic imaging system which allows acquiring and registering low magnification (1 pixel = 5 μm) conventional (bright field or fluorescence) images of entire tissue sections; then it allows tracing (in 3D) the ducts of the mammary gland from adjacent sections, to create a 3D virtual reconstruction of the gland; finally it allows revisiting areas of interest for high resolution (1 pixel = 0.5 μm) imaging and automatic analysis. We illustrate the use of the system for the reconstruction of a small volume of breast tissue.

scholarly journals LUNAR TERRAIN RECONSTRUCTION FROM MULTI-VIEW LROC NAC IMAGES BASED ON SEMI-GLOBAL MATCHING IN OBJECT SPACE

2020 ◽  
Vol XLIII-B3-2020 ◽  
pp. 1177-1183
Author(s):  
L. Ye ◽  
M. Peng ◽  
K. Di ◽  
B. Liu ◽  
Y. Wang
Keyword(s):  
Three Dimensional ◽  
Cost Calculation ◽  
Stereo Pair ◽  
Matching Method ◽  
Global Method ◽  
Object Space ◽  
Three Dimensional Mapping ◽  
Object Based ◽  
Global Matching ◽  
Traditional Image

Abstract. Most of the lunar surface area has been observed from different viewing conditions thanks to the on-orbit work of lunar orbiters, a large amount of images are available for photogrammetric three-dimensional mapping, which is an important issue for lunar exploration. Theoretically, multi-view images contain more information than a single stereo pair and can get better 3D mapping results. In this paper, the semi-global matching method is applied to the object space, and the steps of cost calculation, cost aggregation, and elevation calculation are performed to obtain the three-dimensional coordinates directly. Compared with the traditional image-based semi-global matching method, the object-based semi-global method is more easily extended to multi-view images, which is beneficial for applying multi-view image information. In addition, it does not require steps such as stereo rectification and forward intersection, that is, the overall pipeline is more elegant. Using the LRO NAC images covering Apollo 11 landing area as the experimental data, the result shows that the object-based semi-global matching is competent for the multi-view image matching and the multi-view image result achieves higher accuracy and more details than the single stereo pair. Furthermore, the experimental results of Zhinyu crater data show that this method can also alleviate the uncertainty of the lunar orbiter's positioning to some extent.

scholarly journals B-FGC-Net: A Building Extraction Network from High Resolution Remote Sensing Imagery

2022 ◽  
Vol 14 (2) ◽  
pp. 269
Author(s):  
Yong Wang ◽  
Xiangqiang Zeng ◽  
Xiaohan Liao ◽  
Dafang Zhuang
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Information Integration ◽  
Large Scale ◽  
Spatial Information ◽  
Contextual Information ◽  
Building Extraction ◽  
Remote Sensing Images ◽  
Deep Convolutional Neural Networks ◽  
Level Information

Deep learning (DL) shows remarkable performance in extracting buildings from high resolution remote sensing images. However, how to improve the performance of DL based methods, especially the perception of spatial information, is worth further study. For this purpose, we proposed a building extraction network with feature highlighting, global awareness, and cross level information fusion (B-FGC-Net). The residual learning and spatial attention unit are introduced in the encoder of the B-FGC-Net, which simplifies the training of deep convolutional neural networks and highlights the spatial information representation of features. The global feature information awareness module is added to capture multiscale contextual information and integrate the global semantic information. The cross level feature recalibration module is used to bridge the semantic gap between low and high level features to complete the effective fusion of cross level information. The performance of the proposed method was tested on two public building datasets and compared with classical methods, such as UNet, LinkNet, and SegNet. Experimental results demonstrate that B-FGC-Net exhibits improved profitability of accurate extraction and information integration for both small and large scale buildings. The IoU scores of B-FGC-Net on WHU and INRIA Building datasets are 90.04% and 79.31%, respectively. B-FGC-Net is an effective and recommended method for extracting buildings from high resolution remote sensing images.

scholarly journals Extração de Obstruções Perspectivas de Edifícios ao Longo de Vias Urbanas

2006 ◽  
Vol 33 (2) ◽  
pp. 101
Author(s):  
ANTONIO JULIANO FAZAN ◽  
ALUIR PORFÍRIO DAL POZ ◽  
EDINÉIA APARECIDA DOS SANTOS GALVANIN
Keyword(s):  
High Resolution ◽  
Digital Elevation Model ◽  
Digital Image ◽  
Aerial Images ◽  
Perspective Projection ◽  
Object Space ◽  
Digital Elevation ◽  
Intensity Image ◽  
Elevation Model

This paper presents an approach for extraction of perspective obstructions in high-resolution aerial images caused by the perspective projection of buildings onto adjacent urban ways. The proposed methodology consists in firstly extracting the contours of building roofs and urban ways from an intensity image generated by a conversion of a Digital Elevation Model (DEM). In the following, the polygons representing the roof contours are projected through the perspective bundle onto the respective mean planes of adjacent ways. Intersections between polygons representing roof contours and local segments of adjacent ways allow the extraction of the perspective obstructions in the object-space. The perspective obstruction polygons are finally projected onto the digital image basically using the collinearity equations. The results obtained by methodology allow the verification of its performance and show its potential for extraction of perspective obstructions in high-resolution aerial images.

Three Dimensional materials characterisation at ultrahigh resolution using a position-sensitive atom probe

1990 ◽  
Vol 48 (4) ◽  
pp. 408-409
Author(s):  
R A D Mackenzie ◽  
G D W Smith ◽  
A Cerezo ◽  
T J Godfrey ◽  
J.E. Brown
Keyword(s):  
High Resolution ◽  
Spatial Resolution ◽  
Spatial Information ◽  
Three Dimensional ◽  
Atom Probe ◽  
Sensitive Detector ◽  
Channel Plate ◽  
Position Sensitive ◽  
20 Nm ◽  
Very High

The conventional atom probe field ion microscope permits very high resolution chemical information to be determined with a lateral spatial resolution of typically 2 nm. This spatial resolution is determined by the need to define the analysis area using an aperture. A recent development, the position sensitive atom probe (POSAP), has largely removed this limitation. In a conventional atom probe the ions passing through the aperture, which have come from a circular area of the order of 2 nm in diameter, travel along a long flight path where the mass to charge ratios are determined with high precision. In the position sensitive atom probe the aperture assembly, long flight tube and ion detector (a channel plate) are replaced with a position sensitive detector held at a known distance from the specimen surface. This detector consists of two parts, a channel plate component which permits the flight times (and hence mass to charge ratios) to be determined, and a wedge and strip anode which permits the position of the incoming ion to be calculated. This arrival position corresponds directly to the position on the specimen from which the ion came. The total field of view of the POSAP is a disc approximately 20 nm in diameter. With a conventional atom probe the data acquired during the evaporation sequence can be considered as a core extracted from the specimen, where the average composition as a function of depth is known. The position sensitive atom probe permits us to record data from a much wider core (20 nm rather than 2 nm in diameter), and also to retain the spatial information within the core. As the evaporation proceeds the two dimensional information yielded by the position sensitive detector builds up into a three dimensional block of data. We have, therefore, both chemical and spatial information in three dimensions at very high resolution from the sampled volume of material.

Geometric Modeling and Visualization of Manipulator Workplace

1991 ◽  
Author(s):  
X. Chen ◽  
K. C. Gupta
Keyword(s):  
Geometric Modeling ◽  
Geometric Model ◽  
Three Dimensional ◽  
Extraction Methods ◽  
Boundary Representation ◽  
Three Dimensions ◽  
Contour Method ◽  
Boundary Extraction ◽  
Volumetric Features ◽  
Apparent Contour

Abstract This paper introduces a new and powerful technique to determine the workspace shape of a general n-joint manipulator by utilizing geometric modeling. A complete three-dimensional geometric model of the workspace shape can be generated which is compatible with the computer graphics. The resulting workspace image can be visualized in three-dimensions realistically, manipulated interactively and analyzed for topological and volumetric features. The first part of the paper presents a general discussion of the workspace geometry and boundary representation concept. The remainder of the paper focuses on two workspace boundary extraction methods in a CAD environment: Radial-Slice-Layering method (RSL) and Apparent-Contour method (AC). Several specific examples are presented to illustrate the basic technique.

Sign in / Sign up

Export Citation Format

Share Document