Automatic detection of complex archaeological grazing structures using airborne laser scanning data

Abstract

International audienceThe use of Light Detection And Ranging (LiDAR) for archaeological purposes is becoming more prevalent in order to detect and to document remains located in forested areas. One of the main interests of airborne laser scanning is to put the archaeological information in their context, and to allow a better understanding of the relation between each item and its environment. This concept of archaeological landscape generally results in a too large amount of data to permit a manual analysis. This paper describes an approach for the automatic detection of elementary archaeological grazing structures, found in high concentration in some places of Auvergne (France). These elementary structures are generally connected, creating complex archaeological grazing sets. The detection process is based on the design of a model of an elementary grazing structure. The automatic detection is then carried out, based on the evaluation of the matching degree of each element with the model and on their belonging to complex archaeological grazing structures. The efficiency of the method is tested, by comparison with the manual digitalisation of an expert, on a restricted zone, and the results show that the success rate of the automatic detection reaches higher values than classical template matching approaches. The additional criterion, based on the belonging of each elementary structure to a more complex one, improves the detection success: In a complementary way, this approach offers new opportunities: it is also possible to detect complex structures with a template matching approach, if they contain some simple forms, that can be modelled

Recent Trends and Long-standing Problems in Archaeological Remote Sensing

“…Aqdus, Hanson & Drummond 2012;Cassidy 2009;Challis & Howard 2006;Giardino 2011). More recently, machine learning and computer vision have seen significant recent improvements (Barceló 2010;LeCun, Bengio & Hinton 2015;Szeliski 2010;Schmidhuber 2015;van der Maaten 2006), and have been applied successfully to archaeological remote sensing projects in the past five years (Toumazet et al 2017;Trier & Pilø 2012;Due Trier et al 2016). Machine learning, computer vision, and automation are increasingly seen as instrumental to mobilizing archaeological remote sensing's 'big data', and a priority for research.…”

Using deep neural networks on airborne laser scanning data: Results from a case study of semi‐automatic mapping of archaeological topography on Arran, Scotland

“…Template matching has been used to map burials from optical satellite data (Trier et al, ), and to identify a range of objects including pitfall traps, charcoal burning platforms, and grave mounds in a digital terrain model (DTM) derived from ALS (Schneider, Takla, Nicolay, Raab, & Raab, ; Trier & Pilø, ; Trier & Pilø, ; Trier, Pilø, & Johansen, ; Trier, Zortea, & Tonning, ). Also based on a DTM is an automatic pit filling method based on an inverted DTM to locate mound structures (Freeland et al, ); a combination of curvature estimates, topographic position index, and circular Hough transform to detect prehistoric barrows (Cerrillo‐Cuena, ); a combination of segmentation and template matching to detect grazing structures (Toumazet et al, ); and local contrast in the DTM at three different scales and a random forest classifier to detect burial mounds (Guyot et al, ). A study to detect rectangular enclosures in panchromatic satellite images (Zingman et al, ) concluded that bespoke methods in some cases perform better than using a pre‐trained deep CNN, but at the cost of much longer development time.…”

Object‐based image analysis: a review of developments and future directions of automated feature detection in landscape archaeology

“…Morphological variables used for classification include: Circularity (Davis et al, ; Freeland et al, ; Witharana et al, ) Rectangularity (Zingman et al, ) Area (Davis et al, ; Magnini et al, ; Witharana et al, ) Length and width (Magnini et al, ; Toumazet, Vautier, Roussel, & Dousteyssier, ) Size (Cerrillo‐Cuenca, ; Davis et al, ; Zingman et al, ) Curvature (Cerrillo‐Cuenca, ) Edge detection (Traviglia & Torsello, ; Witharana et al, ; Zingman et al, ) Elevation (Davis et al, ; Guyot et al, ) …”