Line of sight in GIS is a phrase used to described the unimpeded view or access from one point to another point across a terrain or surface. Visibility analysis is a spatial analysis of the portions of a line that are visible as opposed to not visible from a starting point.
Line of sight is used to understand the visible and obstructed (non visible) points in terrain which contains significant changes in elevation (in the form of mountains/hills and valleys).
Uses of Line of Sight in GIS
Line of sight is an important analysis tool in many fields. For example, a developer may use line of sight to determine the best place to site a house on a hillside in order to create the best view of the ocean. The telecommunications industry uses visibility analysis to best site cell phone tower locations which need to have an unobstructed view of other cell phone towers for transmission purposes. For military purposes, line of sight is used to understand optimal positions for battle. In law enforcement, line of sight analysis can be used to canvas a crime scene to search for witnesses.
Creating Line of Sight Using ArcGIS
Doing a line of sight analysis in ArcGIS requires the extension 3D Analyst. Esri has a help page on line of sight. When selecting the line of sight tool using 3D analyst, the user loads in a DEM (digital elevation model) that contains the elevation information required to perform the terrain analysis. Next, the user will need to enter the offset Z values for both the observer and target locations. The offset Z values in the height measurement off the ground that the line of sight occurs at. For example, if the visibility analysis is being performed for a person, it may be set at a height for an average person. For example, if the units are in feet, then the parameter inputted would be 6. If the line of sight analysis is being done to cite a cell phone tower, then the units for the offset Z value would need to reflect the height of the intended cell phone tower. Then use the mouse to click the starting point and drag the mouse over to the ending point for the line of sight visibility. The resulting graphic shows the line of sight (see the image above) which is color coded with green for visible areas and red for non visible areas.
Esri also offers a Linear Line of Sight tool as part of its Military Analyst extension which uses DTED catalogs and surface raster datasets to create a shapefile of visible and non visible points along a line.
Visibility Analysis in QGIS
Line of sight analysis is known as visibility analysis in QGIS. There is a visibility analysis plugin developed by Cagil Seeker in the QGIS plugin repository. The plugin is not part of the official QGIS plugins but can be downloaded from pyqgis.org.
Field of View in GIS
The field of view analysis in GIS is a polygon file that shows the portions of an area that are visible versus no visible across a terrain starting from a given point.
2 thoughts on “Line of Sight in GIS”
Thanks for the interesting article. One thing, though: I have always known the ability to be able to see one point from another over terrain as “intervisibility”. Indeed I was first taught about this in military service thirty years ago, using topographic map interpretation. There was nothing fancy here unless you went to the trouble of creating cross-sections from contour data and working things out accurately. I have heard of and used the phrase “line of sight” but not in this context at all. Is intervisibity a term others recognise/use?
Nice article, visibility is an specific area of interest for me. My graduate thesis in Geography was on the accuracy of the viewshed algorithm using in Esri software (A VIEWSHED ACCURACY ASSESSMENT: COMPARISON OF FIELD-DERIVED AND COMPUTER-DERIVED VIEWSHEDS, Sims, 2010). The body of research I found indicate Line of Sight (LOS) calculations to be more accurate when conducted using Triangular Irregular Networks (TIN) rather than DEMs (Dean, 1997). Esri’s 3D Analyst can be used to perform LOS calculations on TINs.
“Dean (1997) states that the use of TIN models eliminate the inconsistencies in lines of sight derived from grid based DEMs. Vector LOSs can be superimposed over the model rather than raster approximations allowing exact elevation values to be derived as the LOS crosses a facet boundary. While Dean (1997) indicated that viewsheds derived from TINs required greater processing power and time, the results achieved were more accurate.” (Sims, 2010)
The inconsistencies Dean writes about are in the way LOS is calculated on DEMs. The calculation is performed using Breshenham’s vector-to-raster line transformation algorithm. “The algorithm . . . is based on an accumulated error function that keeps track of how far the raster approximation deviates from the original vector line” (Dean 1997, 970).
Viewsheds, refereed to as “field of view” in this article, are basically multiple LOS calculations. I found that all software performing viewsheds performed the calculation on DEMs, no software available at the time of my research was able to perform the viewshed calculation on a TIN. Esri’s software will calculate a viewshed with an input TIN but will convert the TIN to a raster in the background to perform the calculation. In my thesis I propose a methodology for calculating a viewshed on a TIN. Admittedly it was a bruit force method, which is likely not the most elegant solution.
Dean, D. J. 1997. Improving the accuracy of forest viewsheds using triangulated networks and the visual permeability method. Canadian Journal of Forest Research. 27(7): 969-77.
Maloy, M. A., and D. J. Dean. 2001. An accuracy assessment of various GIS-based viewshed delineation techniques. Photogrammetric Engineering and Remotes Sensing. 67 (11): 1293-98
For some other fun reading on the subject I recommend:
Maloy, M. A., and D. J. Dean. 2001. An accuracy assessment of various GIS-based viewshed delineation techniques. Photogrammetric Engineering and Remotes Sensing. 67 (11): 1293-98.
Fisher, P. F. 1991. First experiments in viewshed uncertainty: The accuracy of the viewshed area. Photogrammetric Engineering and Remote Sensing. 57(10): 1321-27.
Fisher, P. F. 1993. Algorithm and implementation uncertainty in viewshed analysis. International Journal of Geographic Information Systems. 7(4) 331-47.
DISCLAIMER: I have not done much work on this topic in several years so new advances may have taken place I am unaware of. Lastly, my response is solely based on my own opinion and academic research.
Comments are closed.