Abstract:Within archaeological prospection, Deep Learning algorithms are developed to detect objects within large remotely sensed datasets. These approaches are generally tested in an (ideal) experimental setting but have not been applied in different contexts or 'in the wild', that is, incorporated in archaeological prospection. This research explores the applicability, knowledge discovery-on both a quantitative and qualitative level-and efficiency gain resulting from employing an automated detection tool called WODAN… Show more
“…This is a more common find in studies that try to detect landforms with deep learning (e.g. Verschoof‐Van der Vaart & Lambers, 2021). Before CNNs can (partly) take over the work of manual mappers in creating geomorphological maps and become part of the workflow for geomorphological mapping, more development and testing of the models, parameters and input is required.…”
Section: Discussionmentioning
confidence: 77%
“…At this stage, semi-automated mapping with CNNs can be seen as an additional tool in the mapping process, as the computed maps can identify areas that are ambiguously defined or may spark the discussion of classification definitions, scale issues and landform dominance. This is a more common find in studies that try to detect landforms with deep learning (e.g Verschoof-Van der Vaart & Lambers, 2021)…”
Geomorphological maps provide information on the relief, genesis and shape of the earth's surface and are widely used in sustainable spatial developments. The quality of geomorphological maps is however rarely assessed or reported, which limits their applicability. Moreover, older geomorphological maps often do not meet current quality requirements and require updating. This updating is time-consuming and because of its qualitative nature difficult to reproduce, but can be supported by novel computational methods. In this paper, we address these issues by (1) quantifying the uncertainty associated with manual geomorphological mapping, (2) exploring the use of convolutional neural networks (CNNs) for semi-automated geomorphological mapping and (3) testing the sensitivity of CNNs to uncertainties in manually created evaluation data.We selected a test area in the Dutch push-moraine district with a pronounced relief and a high variety of landforms. For this test area we developed five manually created geomorphological maps and 27 automatically created landform maps using CNNs. The resulting manual maps are similar on a regional level. We could identify the causes of disagreement between the maps on a local level, which often related to differences in mapping experience, choices in delineation and different interpretations of the legend. Coordination of mapping efforts and field validation are necessary to create accurate and precise maps. CNNs perform well in identifying landforms and geomorphological units, but fail at correct delineation. The human geomorphologist remains necessary to correct the delineation and classification of the computed maps. The uncertainty in the manually created data that are used to train and evaluate CNNs have a large effect on the model performance and evaluation. This also advocates for coordinated mapping efforts to ensure the quality of manually created training and test data. Further model development and data processing are required before CNNs can act as standalone mapping techniques.
“…This is a more common find in studies that try to detect landforms with deep learning (e.g. Verschoof‐Van der Vaart & Lambers, 2021). Before CNNs can (partly) take over the work of manual mappers in creating geomorphological maps and become part of the workflow for geomorphological mapping, more development and testing of the models, parameters and input is required.…”
Section: Discussionmentioning
confidence: 77%
“…At this stage, semi-automated mapping with CNNs can be seen as an additional tool in the mapping process, as the computed maps can identify areas that are ambiguously defined or may spark the discussion of classification definitions, scale issues and landform dominance. This is a more common find in studies that try to detect landforms with deep learning (e.g Verschoof-Van der Vaart & Lambers, 2021)…”
Geomorphological maps provide information on the relief, genesis and shape of the earth's surface and are widely used in sustainable spatial developments. The quality of geomorphological maps is however rarely assessed or reported, which limits their applicability. Moreover, older geomorphological maps often do not meet current quality requirements and require updating. This updating is time-consuming and because of its qualitative nature difficult to reproduce, but can be supported by novel computational methods. In this paper, we address these issues by (1) quantifying the uncertainty associated with manual geomorphological mapping, (2) exploring the use of convolutional neural networks (CNNs) for semi-automated geomorphological mapping and (3) testing the sensitivity of CNNs to uncertainties in manually created evaluation data.We selected a test area in the Dutch push-moraine district with a pronounced relief and a high variety of landforms. For this test area we developed five manually created geomorphological maps and 27 automatically created landform maps using CNNs. The resulting manual maps are similar on a regional level. We could identify the causes of disagreement between the maps on a local level, which often related to differences in mapping experience, choices in delineation and different interpretations of the legend. Coordination of mapping efforts and field validation are necessary to create accurate and precise maps. CNNs perform well in identifying landforms and geomorphological units, but fail at correct delineation. The human geomorphologist remains necessary to correct the delineation and classification of the computed maps. The uncertainty in the manually created data that are used to train and evaluate CNNs have a large effect on the model performance and evaluation. This also advocates for coordinated mapping efforts to ensure the quality of manually created training and test data. Further model development and data processing are required before CNNs can act as standalone mapping techniques.
“…Beyond increasing image contrast for the human perception of archaeological features and natural landforms, lidar visualizations are also utilized to improve automated, or semiautomated, object detection (Davis, 2019;Verschoof-van der Vaart & Lambers, 2022).…”
Section: What Makes a Good Visualization?mentioning
confidence: 99%
“…Beyond increasing image contrast for the human perception of archaeological features and natural landforms, lidar visualizations are also utilized to improve automated, or semiautomated, object detection (Davis, 2019; Verschoof‐van der Vaart & Lambers, 2022). In geomorphology, these procedures are increasingly critical for providing a quantification and recognition of landforms that possess unique morphometric characteristics (Evans, 2012; Jasiewicz & Stepinski, 2013; Lin et al, 2021; Wang et al, 2010).…”
Section: Contemporary Approaches To Lidar Visualizationmentioning
Lidar has become an essential tool for the mapping and interpretation of natural and archaeological features within the landscape. It is also increasingly integrated and visualized within geoarchaeological deposit models, providing valuable topographic and stratigraphic control from the contemporary ground surface downwards. However, there is a wide range of methods available for the visualization of lidar elevation models and a review of existing research suggests that it remains unclear which are most appropriate for geoarchaeological applications. This paper addresses this issue by providing an overview and quantitative evaluation of these techniques with examples from archaeologically resource‐rich alluvial environments. Owing to the relatively low‐relief nature of the terrain within these temperate lowland flood plain environments, the results show that there is a small number of visualization methods that demonstrably improve the detection of geomorphological landforms that can be related to the variable distribution of archaeological resources. More specifically, a combination of Relative Elevation Models combined with Simple Local Relief Models offered an optimal approach that subsequently allows integration with deposit models. Whilst the presented examples are from a flood plain setting, deposit models are pertinent to a range of landscape contexts and the methodology applied here has wider applicability.
“…It has experimented with geospatial data/images (satellite, aerial, lidar), texts, categorical tableau data, point clouds, and other datasets. For instance, one can consider some indicative examples such as the work that has been done on bone classification [1], remote sensing archaeology [2][3][4][5][6][7][8][9][10][11][12], geophysical prospection [13][14][15][16][17], detection of objects in paintings [18], classification of pottery [19], and the 3D reconstruction of heritage buildings [20]. The main reason behind this growing trend, which has been noticed in the last five years in all scientific domains, underlies the nuisance generated when dealing with multivariate analysis of high-volume datasets, which are challenging to process and interpret.…”
Ground penetrating radar (GPR) is a well-established technique used in archaeological prospection and it requires a number of specialized routines for signal and image processing to enhance the data acquired and lead towards a better interpretation of them. Computer-aided techniques have advanced the interpretation of GPR data, dealing with a wide range of operations aiming towards locating, imaging, and diagnosis/interpretation. This article will discuss the novel and recent applications of machine learning (ML) and deep learning (DL) techniques, under the artificial intelligence umbrella, for processing GPR measurements within archaeological contexts, and their potential, limitations, and possible future prospects.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.