Digital Deforestation: DTM Generation with Agisoft Photoscan

Author(s): Matthew Howland; Thomas Levy

Year: 2019


This is an abstract from the "Towards a Standardization of Photogrammetric Methods in Archaeology: A Conversation about 'Best Practices' in An Emerging Methodology" session, at the 84th annual meeting of the Society for American Archaeology.

Applying digital photogrammetry to archaeological sites is a well-known approach. Also fairly common is photogrammetry’s combination with low-altitude aerial photography (LAAP) in order to generate three-dimensional data and produce GIS outputs such as orthophotographs and digital elevation models (DEMs). DEMs are widely useful for a number of spatial analyses and a LAAP-photogrammetry workflow is one of the easiest ways to produce these datasets. However, DEMs produced by these methods often include vegetation, as photogrammetric systems are not able to distinguish between true ground surfaces and the tops of trees. This paper describes a processing technique using point cloud classification that allows users to remove some types of vegetation and architecture in photogrammetric models from consideration when generating a DEM. By doing so, researchers can generate digital terrain models (DTMs), a subset of DEM which factors in only the Earth’s surface and eliminates vegetation and architecture from elevation measurements. These DTMs are better basis for contour generation and various spatial analyses including slope and water flow calculations.

Cite this Record

Digital Deforestation: DTM Generation with Agisoft Photoscan. Matthew Howland, Thomas Levy. Presented at The 84th Annual Meeting of the Society for American Archaeology, Albuquerque, NM. 2019 ( tDAR id: 451426)

Spatial Coverage

min long: 19.336; min lat: 41.509 ; max long: 53.086; max lat: 70.259 ;

Record Identifiers

Abstract Id(s): 22942