Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands
Computer-aided methods for the automatic detection of archaeological objects are needed to cope with the ever-growing set of largely digital and easily available remotely sensed data. In this paper, a promising new technique for the automated detection of multiple classes of archaeological objects i...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ubiquity Press
2019-03-01
|
| Series: | Journal of Computer Applications in Archaeology |
| Subjects: | |
| Online Access: | https://journal.caa-international.org/articles/32 |
| _version_ | 1818941924897193984 |
|---|---|
| author | Wouter Baernd Verschoof-van der Vaart Karsten Lambers |
| author_facet | Wouter Baernd Verschoof-van der Vaart Karsten Lambers |
| author_sort | Wouter Baernd Verschoof-van der Vaart |
| collection | DOAJ |
| description | Computer-aided methods for the automatic detection of archaeological objects are needed to cope with the ever-growing set of largely digital and easily available remotely sensed data. In this paper, a promising new technique for the automated detection of multiple classes of archaeological objects in LiDAR data is presented. This technique is based on R-CNNs (Regions-based Convolutional Neural Networks). Unlike normal CNNs, which classify the entire input image, R-CNNs address the problem of object detection, which requires correctly localising and classifying (multiple) objects within a larger image. We have incorporated this technique into a workflow, which enables the preprocessing of LiDAR data into the required data format and the conversion of the results of the object detection into geographical data, usable in a GIS environment. The proposed technique has been trained and tested on LiDAR data gathered from the central part of the Netherlands. This area contains a multitude of archaeological objects, including prehistoric barrows and Celtic fields. The initial experiments show that we are able to automatically detect and categorise these two types of archaeological objects and thus proof the added value of this technique. |
| first_indexed | 2024-12-20T07:03:16Z |
| format | Article |
| id | doaj.art-e9967ae194714d218afea34274d4182b |
| institution | Directory Open Access Journal |
| issn | 2514-8362 |
| language | English |
| last_indexed | 2024-12-20T07:03:16Z |
| publishDate | 2019-03-01 |
| publisher | Ubiquity Press |
| record_format | Article |
| series | Journal of Computer Applications in Archaeology |
| spelling | doaj.art-e9967ae194714d218afea34274d4182b2022-12-21T19:49:09ZengUbiquity PressJournal of Computer Applications in Archaeology2514-83622019-03-012110.5334/jcaa.3215Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the NetherlandsWouter Baernd Verschoof-van der Vaart0Karsten Lambers1Leiden UniversityLeiden UniversityComputer-aided methods for the automatic detection of archaeological objects are needed to cope with the ever-growing set of largely digital and easily available remotely sensed data. In this paper, a promising new technique for the automated detection of multiple classes of archaeological objects in LiDAR data is presented. This technique is based on R-CNNs (Regions-based Convolutional Neural Networks). Unlike normal CNNs, which classify the entire input image, R-CNNs address the problem of object detection, which requires correctly localising and classifying (multiple) objects within a larger image. We have incorporated this technique into a workflow, which enables the preprocessing of LiDAR data into the required data format and the conversion of the results of the object detection into geographical data, usable in a GIS environment. The proposed technique has been trained and tested on LiDAR data gathered from the central part of the Netherlands. This area contains a multitude of archaeological objects, including prehistoric barrows and Celtic fields. The initial experiments show that we are able to automatically detect and categorise these two types of archaeological objects and thus proof the added value of this technique.https://journal.caa-international.org/articles/32Remote sensingObject detectionR-CNNMachine learning |
| spellingShingle | Wouter Baernd Verschoof-van der Vaart Karsten Lambers Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands Journal of Computer Applications in Archaeology Remote sensing Object detection R-CNN Machine learning |
| title | Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands |
| title_full | Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands |
| title_fullStr | Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands |
| title_full_unstemmed | Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands |
| title_short | Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands |
| title_sort | learning to look at lidar the use of r cnn in the automated detection of archaeological objects in lidar data from the netherlands |
| topic | Remote sensing Object detection R-CNN Machine learning |
| url | https://journal.caa-international.org/articles/32 |
| work_keys_str_mv | AT wouterbaerndverschoofvandervaart learningtolookatlidartheuseofrcnnintheautomateddetectionofarchaeologicalobjectsinlidardatafromthenetherlands AT karstenlambers learningtolookatlidartheuseofrcnnintheautomateddetectionofarchaeologicalobjectsinlidardatafromthenetherlands |