Aspekter kring höjd/3D-data i geoförsörjningsprocessen
Publish date: 2014-10-17
Report number: FOI-R--3950--SE
Pages: 39
Written in: Swedish
Keywords:
- Monte Carlo-simulation
- error propagation
- viewshed analysis
- DGIWG
- ESM
- sensor Models
- accuracy
- visualization
- compression
- datasets
- scenario
- radar
- satellites.
Abstract
Viewshed analysis using elevation data is often a desirable function for planning and realization of a military operation. The accuracy in the result from a viewshed analysis depends highly on the accuracy of the input data. Even small errors in the input data could have a great influence on the analysis. Besides the error in height, there are also planar uncertainties, partly from height data and partly from the observers' position. These faults could be the difference between if the position of the observer ends up behind an object that blocks the sight or on a spot where there will be a free line of sight. To investigate how it is possible to considering the uncertainties, a method for analyze error propagation that builds on Monte Carlo simulation has been produced. With this method, a large number of calculations are made with different values for the parameters such as error in the height in the areas of vegetation, error in the height in areas of open spaces and uncertainties in plane. The result from the simulation is analyzed statistically to see how the variations in the parameters are affecting the result. The result from the method is a viewshed analysis where the likelihood for free line of sight could be more balanced then just line of sight or no line of sight. Then it is also important to present the information for the operator in such way so the operator couldn't misunderstand what the information is really representing. To increase the understanding of what the result from the field analysis implies and to get an opportunity to discuss how heterogeneous geographic information (that already exist in the Swedish Armed forces today) could be used at a military action, a scenario has been produced for a coming demonstration. In this way, the usefulness of the analysis could be put into the context and the importance of taking the errors into account could be shown, but also provide an opportunity to discuss what additional capabilities that may be needed along with those that already exist in today's GIS system. Today, there are standards that have been produced with the aim of improving the quality assurance of increasingly high-resolution elevation data that are available today. In this report, we are looking at some of the standards that exist in this area. To get a survey of how it looks today and what is going on in some areas of interest for the FMV project SGIF (Rapid geographic information management), a limited outlook have been made for areas such as compression, databases, large data sets, presentation of 2D and 3D data and radar satellites.