An automated approach for constructing road network graph from multispectral images

2012 ◽  
Author(s):  
Weihua Sun ◽  
David W. Messinger
Keyword(s):  
Road Network ◽  
Multispectral Images ◽  
Network Graph

scholarly journals OPTIMAL CONTROL WITH THE DYNAMIC CHANGE OF THE STRUCTURE OF THE ROAD NETWORK

Transport ◽  
2014 ◽  
Vol 29 (1) ◽  
pp. 36-42 ◽  
Author(s):  
Zsuzsanna Bede ◽  
Tamás Péter
Keyword(s):  
Optimal Control ◽  
Road Network ◽  
Road Traffic ◽  
Dynamic Change ◽  
Time Of Day ◽  
Traffic Density ◽  
Special Method ◽  
Network Graph ◽  
The Road ◽  
Traffic Systems

Optimization of traffic on a large public road network is a complex task. Reversible direction lane theory is an interesting and special method within this subject. This can completely support the fluctuation or alteration of main congested directions existing in the traffic dynamics (time of day, seasonal, etc.) on the existing road surfaces. In such case, certain subsystems of the main network cease to exist, and subsystems working with new connections take their place. This type of routing therefore changes the structure of the system ‘in an optimal direction’, but many practical and safety questions arise. The authors have examined the modelling of a Reversible Lane System (RLS) created based on a simple part of a road network, which is segmented into elements. Functions of each network element and contacts between them cease operating in the course of such change while new contacts and new function elements are activated instead. The article presents the mathematical modelling of the problem. It points out the fundamental questions of the structure change and exemplifies the above using a simple example. The authors studied a general mathematical model describing the RLS. They examined the availability of the optimal control in a sample network depending on the traffic density, using a new principle, which responds to the dynamic change of the structure of the network graph. It can be shown, that the results from the model are in harmony with the real traffic values based on measurements made in road traffic systems working with RLS.

scholarly journals Characterization of Jos City Road Network, Nigeria

2014 ◽  
Vol 46 (2) ◽  
pp. 177
Author(s):  
Olumide Akinwumi Oluwole ◽  
Nwanret Gideon Daful
Keyword(s):  
Road Network ◽  
City Centre ◽  
Network Graph ◽  
Road Density ◽  
Graph Theoretic ◽  
The Road ◽  
Study Results ◽  
Cyclomatic Number ◽  
Index Value ◽  
Alpha Index

ἀe performance of road network depends on its topological characteristics which help to deḀne its connectiv-ity. ἀis paper analyses the topological characteristic of Jos city road network and its bearing on traᴀc ᰀow situations. Simple graph theoretic measures oᬀered the framework on which the problem was approached. ἀe study requires the abstraction and analysis of the topological structure by selection of certain variables relating to the road connectivity. ἀese include the Beta, Gamma and Alpha index, the PI, Cyclomatic number, and the spread and density of the network. Information on these variables was obtained through the use of vector data model to abstract the road network graph from the Quick-bird satellite imagery used for the study. Results of the Ḁndings reveal that, the road network of Jos City Centre as a whole have achieved an average level of connectivity, showing Beta index values of 1.4049, Gamma index value of 47.06%, Alpha index Value of 20.63%; and a pi and cyclomatic number of 24.74 and 165 respectively, the spread of the network is moderate exhibiting a value of 23, even though some areas have more concentration of roads than the others; and has a road density of 52 links per km2. Based on these Ḁndings, the need for construction of new roads is imperative so as to improve the eᴀciency of connectivity and accessibility within the city.

scholarly journals FEATURES OF THE CLASSIFICATION OF THE STREET-ROAD NETWORK WHEN CONSTRUCTING THE ROAD GRAPH IN THE TRANSPORT MODEL

2021 ◽  
Vol 7 (2) ◽  
pp. 268-283
Author(s):  
V. N. Myachin ◽  
◽  
K. S. Borovikova ◽  
D. P. Krivtsov ◽  
◽  
...  
Keyword(s):  
Russian Federation ◽  
Road Network ◽  
Model Building ◽  
Transport Model ◽  
Network Graph ◽  
Normative Documents ◽  
The Road ◽  
The Russian Federation ◽  
Transport Problems

Constructing transport models is a relevant tool for solving various transport problems. The article discusses one of the important stages of creating a transport model — building a graph of a street-road network. The examples of the graphs of the street-road network of cities developed by the authors are presented. Special attention is paid to the features of road classification when developing a graph. The analysis of normative documents, in accordance with which a class is assigned to roads and streets in the Russian Federation, is carried out. Three ways of developing a road network graph are proposed. Methods for constructing a graph using data on the street-road network from the open cartographic source OSM are described in more detail. The main problems of applying each of the methods are outlined. The main problem is the discrepancy between the classification of roads from the OSM and the classification of roads of the Russian Federation adopted in the regulatory documents. Variants of simplifying the construction of the road network graph in the transport model are suggested.

scholarly journals RoadTagger: Robust Road Attribute Inference with Graph Neural Networks

2020 ◽  
Vol 34 (07) ◽  
pp. 10965-10972
Author(s):  
Songtao He ◽  
Favyen Bastani ◽  
Satvat Jagwani ◽  
Edward Park ◽  
Sofiane Abbar ◽  
...  
Keyword(s):  
Neural Networks ◽  
Receptive Field ◽  
Spatial Correlation ◽  
Satellite Imagery ◽  
Road Network ◽  
Network Graph ◽  
The Road ◽  
Road Type ◽  
On The Road ◽  
Graph Neural Networks

Inferring road attributes such as lane count and road type from satellite imagery is challenging. Often, due to the occlusion in satellite imagery and the spatial correlation of road attributes, a road attribute at one position on a road may only be apparent when considering far-away segments of the road. Thus, to robustly infer road attributes, the model must integrate scattered information and capture the spatial correlation of features along roads. Existing solutions that rely on image classifiers fail to capture this correlation, resulting in poor accuracy. We find this failure is caused by a fundamental limitation – the limited effective receptive field of image classifiers.To overcome this limitation, we propose RoadTagger, an end-to-end architecture which combines both Convolutional Neural Networks (CNNs) and Graph Neural Networks (GNNs) to infer road attributes. Using a GNN allows information to propagate on the road network graph and eliminates the receptive field limitation of image classifiers. We evaluate RoadTagger on both a large real-world dataset covering 688 km2 area in 20 U.S. cities and a synthesized dataset. In the evaluation, RoadTagger improves inference accuracy over the CNN image classifier based approaches. In addition, RoadTagger is robust to disruptions in the satellite imagery and is able to learn complicated inductive rules for aggregating scattered information along the road network.

scholarly journals Identificación De Redes De Caminos Utilizando Imágenes Digitales Multi-Espectrales Y Pancromáticas. El Caso De Los Ramales Del Camino Real De Tierra Adentro En San Juan Del Rí­o

Xihmai ◽  
2014 ◽  
Vol 9 (18) ◽  
Author(s):  
Fernando López Aguilar ◽  
Pedro López Garcí­a
Keyword(s):  
Road Network ◽  
Spectral Characteristics ◽  
Digital Processing ◽  
Multispectral Images ◽  
Road Extraction ◽  
Edge Information ◽  
Spatial Features ◽  
Camino Real ◽  
Palabras Clave ◽  
Medium Resolution

Resumen Los métodos de clasificación para la identificación de caminos usando imágenes de satélite de alta resolución se basan en las caracterí­sticas espectrales de estas; sin embargo, en la actualidad, el procesamiento digital permite integrar las caracterí­sticas espectrales con las caracterí­sticas espaciales al fusionar imágenes pancromáticas de alta resolución con imágenes multiespectrales de resolución media. Este procesamiento toma el nombre de Pansharpening. Posteriormente, la imagen producto de la fusión es procesada utilizando algoritmos de clasificación para discriminar los caminos o trazos lineales de otros objetos como pueden ser ciudades, parcelas, cuerpos de agua, etc. Los caminos clasificados pueden ser subsecuentemente re-clasificados utilizando la información de los bordes para eliminar objetos que no corresponden a caminos. Palabras clave: Mapas, redes de caminos, imágenes digitales multiespectrales, pancromáticas Abstract Classification methods for road extraction are based in the spectral characteristics of the images. However, actually, the digital processing allows the integration of the spectral characteristics with the spatial features when combining high resolution panchromatic images with medium resolution multispectral images. This processing is known as Pansharpening. The resulting image is classified to discriminate between roads and other objects, like cities, water bodies and crops, between others. The identified roads are then segmented and re-classified using the edge information to eliminate the features that do not correspond to roads.   Keywords: Maps, Road network, Multispectral and Panchromatic Digital Images

Sign in / Sign up

Export Citation Format

Share Document