Date of Graduation
5-2023
Document Type
Thesis
Degree Name
Bachelor of Arts
Degree Level
Undergraduate
Department
Anthropology
Advisor/Mentor
Vining, Benjamin
Committee Member/Reader
Stoner, Wesley
Committee Member/Second Reader
Tullis, Jason
Abstract
This study represents a first attempt at applying geographic object-based image analysis (GEOBIA) for the detection of extant archaeological sites in the Chicama River Valley located in northern coastal Peru. This region is characterized by the presence of mound-like archaeological sites dispersed throughout large-scale agricultural fields. High-resolution multispectral imagery from Planet Labs was used in this study because of its high spatial and temporal resolution and availability at no cost through Planet’s Education and Research Program. The main goal for this study was to explore GEOBIA as an approach to archaeological site detection and assess the feasibility of developing exportable classification rule-sets which ideally could be applied to cultural heritage monitoring. Four classification rule-sets were developed and calibrated using four distinct, judgmentally-selected 2 x 2 km sub-areas within the study area. Each rule-set was then applied to the full study area. The preliminary classifications resulting from rule-set calibration were generally promising, but the final classifications for the full study region exhibited major inaccuracies. However, insight gained from the rule-set development and classification processes can inform future studies of methodological and region-specific challenges and of potential criteria for future GEOBIA attempts.
Keywords
archaeology; GIS; image analysis; cultural heritage
Citation
Craig, C. (2023). Geographic Object-Based Image Analysis for Archaeological Site Detection: A Case Study for the Chicama Valley, Peru. Anthropology Undergraduate Honors Theses Retrieved from https://scholarworks.uark.edu/anthuht/10
