Machine Learning R-CNN Identification of the Entirety of the Southwest Regional Road Network
Author(s): Kelsey Reese; Sean Field; Robert Weiner
Year: 2023
Summary
This is an abstract from the "SAA 2023: Individual Abstracts" session, at the 88th annual meeting of the Society for American Archaeology.
The United States Geologic Survey is intermittently releasing publicly available 1-meter resolution lidar of the contiguous United States through the 3-Dimensional Elevation Project. Over the past several years, large sections of lidar across southeast Utah, southwest Colorado, New Mexico, and small portions of Arizona have been released—creating an unprecedented perspective of the Ancestral Pueblo footprint still embedded across the U.S. Southwest at a landscape-wide scale. Examining the lidar imagery reveals a vast network of linear and circular road-like features far more extensive than those typically attributed to the Chaco regional network, and many associated with communities primarily occupied beyond the height of the Chaco Period. This poster will present results from a Regional Convolutional Neural Network created to systematically identify all road features across the greater Four Corners region.
Cite this Record
Machine Learning R-CNN Identification of the Entirety of the Southwest Regional Road Network. Kelsey Reese, Sean Field, Robert Weiner. Presented at The 88th Annual Meeting of the Society for American Archaeology. 2023 ( tDAR id: 475144)
This Resource is Part of the Following Collections
Keywords
Geographic Keywords
North America: Northern Southwest U.S.
Spatial Coverage
min long: -123.97; min lat: 37.996 ; max long: -101.997; max lat: 46.134 ;
Record Identifiers
Abstract Id(s): 37609.0