Pixel versus object — A comparison of strategies for the semi-automated mapping of archaeological features using airborne laser scanning data (original) (raw)
Related papers
AIRBORNE LASER SCANNING AND IMAGE PROCESSING TECHNIQUES FOR ARCHAEOLOGICAL PROSPECTION
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XL-5, 2014
Aerial photography was, for decades, an invaluable tool for archaeological prospection, in spite of the limitation of this method to deforested areas. The airborne laser scanning (ALS) method can be nowadays used to map complex areas and suitable complement earlier findings. This article describes visualization and image processing methods that can be applied on digital terrain models (DTMs) to highlight objects hidden in the landscape. Thanks to the analysis of visualized DTM it is possible to understand the landscape evolution including the differentiation between natural processes and human interventions. Different visualization methods were applied on a case study area. A system of parallel tracks hidden in a forest and its surroundings-part of old route called "Devil's Furrow" near the town of Sázava was chosen. The whole area around well known part of Devil's Furrow has not been prospected systematically yet. The data from the airborne laser scanning acquired by the Czech Office for Surveying, Mapping and Cadastre was used. The average density of the point cloud was approximately 1 point/m 2. The goal of the project was to visualize the utmost smallest terrain discontinuities, e.g. tracks and erosion furrows, which some were not wholly preserved. Generally we were interested in objects that are clearly not visible in DTMs displayed in the form of shaded relief. Some of the typical visualization methods were tested (shaded relief, aspect and slope image). To get better results we applied image-processing methods that were successfully used on aerial photographs or hyperspectral images in the past. The usage of different visualization techniques on one site allowed us to verify the natural character of the southern part of Devil's Furrow and find formations up to now hidden in the forests.
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017
The archaeological heritage is non-renewable, and any invasive research or other actions leading to the intervention of mechanical or chemical into the ground lead to the destruction of the archaeological site in whole or in part. For this reason, modern archeology is looking for alternative methods of non-destructive and non-invasive methods of new objects identification. The concept of aerial archeology is relation between the presence of the archaeological site in the particular localization, and the phenomena that in the same place can be observed on the terrain surface form airborne platform. One of the most appreciated, moreover, extremely precise, methods of such measurements is airborne laser scanning. In research airborne laser scanning point cloud with a density of 5 points/sq. m was used. Additionally unmanned aerial vehicle imagery data was acquired. Test area is located in central Europe. The preliminary verification of potentially microstructures localization was the creation of digital terrain and surface models. These models gave an information about the differences in elevation, as well as regular shapes and sizes that can be related to the former settlement/sub-surface feature. The paper presents the results of the detection of potentially sub-surface microstructure fields in the forestry area.
Semi-automatic mapping of cultural heritage from airbone laser scanning data
2016
This chapter describes semi-automatic methods for detailed mapping of some types of archaeological structure from airborne laser scanning (ALS) data in Norway. The archaeological structures include point or circular features like: (1) pitfall traps in hunting systems, (2) charcoal burning pits in iron extraction sites, (3) grave mounds, either individual or in grave fields, and (4) charcoal kilns. We report on the results of using the methods for detailed mapping of Norway’s largest Viking grave field at Vang, Oppdal municipality in Sor-Trondelag County; and for detailed mapping of charcoal kilns at Lesja iron works in Oppland County. The perceived success of using automatic detection methods depend on many factors. The most important seems to be how well the archaeological structures of interest stand out from the surrounding terrain in the digital terrain model of the ALS ground points. We have experienced that archaeological pits in the form of pitfall traps in hunting systems an...
Journal of Archaeological Science 39, 698–703, 2011
In this paper we report the results of an experiment with automated landform delineation and classification from digital elevation models (DEMs) using object-based image analysis (OBIA). Archaeologists rely on accurate and detailed geomorphological maps to predict and interpret the location of archaeological sites. However, they have been using high-resolution DEMs primarily for visual interpretation and expert-judgement classification of landform. OBIA can perform these classifications much faster and in a more objective fashion. The method was tested on a study area in the south east of the Netherlands. It is concluded that OBIA is a suitable technique for quick and objective delineation of landform, but needs an improved conceptual framework adapted to the local situation and archaeological questions to better identify and interpret the derived landform objects.
Journal of Archaeological Science, 2011
"Airborne Light Detection and Ranging (LiDAR) is a quite recent (mid-1990s) remote sensing technique used to measure terrain elevation. Recent studies have examined the possibility of using LiDAR in archaeological investigations to map and characterize earthworks, to capture features that may be indistinguishable on the ground and to aid the planning of archaeological excavation campaigns. Despite the great potential of LiDAR in archaeology, also linked to its unique capability to penetrate vegetation canopies and identify archaeological earthworks and remains even under dense vegetation cover, the use of airborne laser scanning data encounters serious challenges. Data filtering and processing as well as pattern extraction, classification of terrain information from raw LiDAR data is still a challenging ongoing research. In this paper, we present the data processing chain along with the threshold-based algorithm we devised for the classification of ground and non-ground points and for the detection of archaeological features. The classification of laser data was performed using a strategy based on a set of “filtrations of the filtrate”. Appropriate criteria for the classification and filtering were set to gradually refine the intermediate results in order to obtain the vegetation heights and to discriminate between canopy, understory and micro-topographic relief of archaeological interest. We selected sample areas within two abandoned medieval settlements in Southern Italy characterized by the presence of low and heterogeneous herbaceous cover and complex topographical and morphological features, which make the identification of archaeological features really complex. Results from our investigations pointed out that the applied data processing enables the detection of micro-topographic relief in sparsely as well as in densely vegetated areas. The most important facts to cope with different environmental situations are mainly linked with (i) the resolution of the acquired data set and (ii) the data acquisition and processing chain specifically devised for archaeological purposes."
2013
In this study, detection success rates were evaluated for cultural remains that were detected manually based on interpretation of digital terrain models (DTM) derived from airborne laser scanning data and with a resolution of 1, 5 and 10 points m À2 . The group of cultural remains included charcoal kilns, charcoal pits, hollow-roads, various pits, house foundations, tar kilns, grave mounds and pit-falls. The effects on the interpretation success of different types of cultural remains and their physical properties were studied: size, shape and elevation difference showing that the detection success rates varied considerably. The main tendency was that large cultural remains with clear geometrical shape (ovals and circles) and large elevation difference were much more successfully detected and classified compared to the smaller ones, especially those without a clear geometrical shape. The study also showed that it was the identification of the larger structures which profited most from an increased resolution of the DTM, and it was of no help to increase resolution in order to improve the identification of the irregularly shaped cultural remains.
Interpreting archaeological topography: lasers, 3D data, observation, visualisation and applications
2013
Airborne Laser Scanning (ALS), or lidar, is an enormously important innovation for data collection and interpretation in archaeology. The application of archaeological 3D data deriving from sources including ALS, close-range photogrammetry and terrestrial and photogrammetric scanners has grown exponentially over the last decade. Such data present numerous possibilities and challenges, from ensuring that applications remain archaeologically relevant, to developing practices that integrate the manipulation and interrogation of complex digital datasets with the skills of archaeological observation and interpretation. This volume addresses the implications of multi-scaled topographic data for contemporary archaeological practice in a rapidly developing field, drawing on examples of ongoing projects and reflections on best practice.Twenty papers from across Europe explore the implications of these digital 3D datasets for the recording and interpretation of archaeological topography, whether at the landscape, site or artefact scale. The papers illustrate the variety of ways in which we engage with archaeological topography through 3D data, from discussions of its role in landscape archaeology, to issues of context and integration, and to the methodological challenges of processing, visualisation and manipulation. Critical reflection on developing practice and implications for cultural resource management and research contextualize the case studies and applications, illustrating the diverse and evolving roles played by multi-scalar topographic data in contemporary archaeology.
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLI-B5, 2016
This paper deals with the non-destructive documentation of the "Radkov" (Svitavy district, Czech Republic) archaeological site. ALS, GPR and land survey mapping will be used for the analysis. The fortified hilltop settlement "Radkov" is an immovable historical monument with preserved relics of anthropogenic origin in relief. Terrain reconnaissance can identify several accentuated objects on site. ALS enables identification of poorly recognizable archaeological objects and their contexture in the field. Geophysical survey enables defunct objects identification. These objects are hidden below the current ground surface and their layout is crucial. Land survey mapping provides technical support for ALS and GPR survey. It enables data georeferencing in geodetic reference systems. GIS can then be used for data analysis. M. Cejpová and J. Němcová have studied this site over a long period of time. In 2012 Radkov was surveyed using ALS in the project "The Research of Ancient Road in Southwest Moravia and East Bohemia". Since 2015 the authors have been examining this site. This paper summarises the existing results of the work of these authors. The digital elevation model in the form of a grid (GDEM) with a resolution 1 m of 2012 was the basis for this work. In 2015 the survey net, terrain reconnaissance and GPR survey of two archaeological objects were done at the site. GDEM was compared with these datasets. All datasets were processed individually and its results were compared in ArcGIS.